making equality effective

Making Equality Effective: The role of proactive measures

EUROPEAN NETWORK OF LEGAL EXPERTS IN THE FIELD OF GENDER EQUALITY

Sandra Fredman

European Commission Directorate-General for Employment, Social Affairs and Equal Opportunities

Unit EMPL/G/2 Theme

Equality, Action against Discrimination: Legal Questions

This publication has been commissioned by the European Commission under the framework programme PROGRESS (Decision 1672/2006/EC of the European Parliament and the Council, OJ L 315/1 of 15.11.2006). For more information on PROGRESS see: http://ec.europa.eu/employment_social/progress/index_en.htm Production: Susanne Burri Frans van Eck Sandra Fredman Titia Hijmans van den Bergh Titia Kloos Christopher McCrudden Simone van der Post Manuscript completed in December 2009. The information contained in this report reflects, as far as possible, the state of affairs on 31 August 2009. The contents of this document do not necessarily reflect the opinion or position of the European Commission, Directorate-General for Employment, Social Affairs and Equal Opportunities. Neither the European Commission nor any person acting on its behalf is responsible for the use which might be made of the information in this publication.

Making Equality Effective: The role of proactive measures i

Table of Contents Members of the European Network of Legal Experts in the Field of Gender Equality ii Making Equality Effective: The role of proactive measures 1 Sandra Fredman Executive summary 1 1. The pure complaints-led model 1 2. Modified complaints-led models 2 3. Proactive measures 3 3.1 Eliminating unlawful discrimination: the case of equal pay 3 3.2 Preventing inequality: impact assessment 4 3.3 Promoting equality: structural change 4 4. Ingredients of a proactive model 5 4.1 Responsibility 5 4.2 Participation 6 4.3 Monitoring 6 4.4 Enforcement and compliance 7 5. Conclusion 8 I Introduction 9 II The complaints-led model 11 III Modified complaints model 15 1. Minimising the burden on the individual 16 2. Modifying the adjudicative body 18 3. Impact of modified complaints models 22 4. Overall assessment of complaints model and modifications 25 IV Proactive models 28 1. The nature of proactive measures 29 1.1 Making anti-discrimination law more effective: the case of equal pay 30 1.2 Preventing inequality: impact assessment 34 1.3 Promoting equality 36 1.3.1 Quotas 36 1.3.1.1 Lawfulness of quotas 36 1.3.1.2 Public employment 39 1.3.1.3 Quotas in the legislature 40 1.3.1.4 Judiciary and other decision-making bodies 41 1.3.1.5 Private sector 42 1.3.1.6 Effectiveness 42 1.3.2 Family-friendly measures 43 1.3.2.1 General 43 1.3.2.2 Child-care 45 1.3.2.3 Fathers 46 2. Structure of the proactive duties 49 2.1 Responsibility for taking proactive measures? 49 2.2 Consultation 52 2.2.1 The subject matter 53 2.2.2 The parties 54 2.2.3 Selection 58 2.2.4 Is consultation binding? 59 2.2.5 Assessment 60

ii Making Equality Effective: The role of proactive measures

2.3 Monitoring 61 2.3.1 Confidentiality and data protection 65 2.4 Enforcement 66 2.4.1 Reporting to Parliament 67 2.4.2 Equality bodies 68 2.4.3 Individual complaints 70 2.4.4 Effectiveness 72 V Conclusion 75 Annex I: Questionnaire 77 Annex II: Bibliography 83

Making Equality Effective: The role of proactive measures iii

Members of the European Network of Legal Experts in the Field of Gender Equality

Co-ordinator: Susanne Burri, Utrecht University, the Netherlands Assistant co-ordinator: Hanneke van Eijken, Utrecht University, the Netherlands Executive Committee: Sacha Prechal, Utrecht University, the Netherlands Christopher McCrudden, Oxford University, the United Kingdom Hélène Masse-Dessen, Barrister at the Conseil d’Etat and Cour de Cassation, France Susanne Burri, Utrecht University, the Netherlands National experts: Anna Sporrer (Austria), Association for Women’s Access to Justice Jean Jacqmain (Belgium), Free University of Brussels, Faculty of Law Genoveva Tisheva (Bulgaria), Bulgarian Gender Research Foundation Evangelia Lia Eftstratiou-Georgiades (Cyprus), Advocate Kristina Koldinská (Czech Republic), Charles University, Faculty of Law Ruth Nielsen (Denmark), Copenhagen Business School, Law Department Anneli Albi (Estonia), University of Kent, Kent Law School Kevät Nousiainen (Finland) Helsinki University, Faculty of Law Sylvaine Laulom (France), University Jean Monnet, Saint-Etienne, Faculty of Law Beate Rudolf (Germany), Free University of Berlin, Department of Law Sophia Koukoulis-Spiliotopoulos (Greece), Attorney at law Csilla Kollonay Lehoczky (Hungary), Eötvös Loránd University, Department of Labour and Social Law Herdís Thorgeirsdóttir (Iceland), Bifrost University, Faculty of Law Frances Meenan (Ireland), Barrister, Four Courts, Law Library, Dublin Simonetta Renga (Italy), Ferrara University, Faculty of Economics Kristīne Dupate (Latvia), University of Latvia, Faculty of Law Nicole Mathé (Liechtenstein), University of Vienna Tomas Davulis (Lithuania), Vilnius University, Faculty of Law Anik Raskin (Luxembourg), National Women’s Council of Luxembourg Peter G. Xuereb (Malta), University of Malta, Faculty of Law Rikki Holtmaat (the Netherlands), University of Leiden, Faculty of Law Helga Aune (Norway), University of Oslo, Faculty of Law Eleonora Zielinska (Poland), University of Warsaw, Faculty of Law and Administration Maria do Rosário Palma Ramalho (Portugal), University of Lisbon, Faculty of Law Roxana Teşiu (Romania), Independent legal advisor Zuzana Magurová (Slovakia), Slovak Academy of Sciences, Institute of State and Law Tanja Koderman Sever (Slovenia), Independent legal advisor Berta Valdes (Spain) University Castilla-La Mancha, Faculty of Law Ann Numhauser-Henning (Sweden), Lund University, Faculty of Law Aileen McColgan (the United Kingdom), King’s College London Ad hoc experts: Dagmar Schiek, University of Leeds, the United Kingdom Christa Tobler, University of Leiden, the Netherlands and University of Basel, Switzerland

Making Equality Effective: The role of proactive measures 1

Making Equality Effective: The role of proactive measures

Sandra Fredman*

EXECUTIVE SUMMARY Gender inequality across Europe remains elusive. Figures from 2007 show that across Europe women earn on average 17.4 % less than men. Although women’s participation in the paid labour force is increasing, labour markets continue to be highly segregated, with women clustering in lower paid sectors or grades. Moreover, the increased participation of women in the labour market is largely characterised by a high proportion of part-time work. This reflects the fact that women remain primarily responsible for child-care, care of the elderly and the disabled. In addition, women are under-represented in national Parliaments and other decision-making bodies.1 Such inequalities persist despite a deep commitment at EU level to the achievement of gender equality, as well as a sophisticated framework of anti-discrimination laws across the EU member states. The tenacity of gender inequality suggests the need to re-examine the methods used to achieve real and substantive equality between men and women. The traditional approach has been to rely on an individual complaints model of adjudication. But a range of new approaches are emerging, which aim at institutional change through proactive measures to promote equality. This report assesses these models in the light of current practice among the Member States of the EU and EEA. For the purposes of this report, experts from the 27 EU Member States and Iceland, Liechtenstein and Norway filled in the questionnaire at Appendix 1. 1. The pure complaints-led model The pure complaints-led model requires an individual victim to bring a complaint to court to establish a breach of her right not to be discriminated against. When it functions well, this approach is an important avenue of redress for individuals. Unlike proactive measures, it clearly defines who has rights and remedies in cases of gender discrimination. Substantive anti-discrimination law, such as the principle of indirect discrimination, has the potential to address disparate impact as well as unequal treatment. Moreover, the complaints-led model does not require official consent or action for the individual to challenge discrimination affecting her. However, it is clear that it is not sufficient to rely solely on a complaints-led model. Although the possibility of judicial recourse is universally provided, there are remarkably few claims compared to the scale of gender inequality revealed by the figures above. In many Member States, the number of gender discrimination cases before the courts has remained in single figures. Even where there is somewhat greater usage, this significantly under-represents the scale of gender inequality.

* Professor of Law, Oxford University. I am grateful to the country experts for their helpful and

illuminating reports, as well as to Susanne Burri, Christopher McCrudden, Simone van der Post and Tolga Yalkin for their assistance on this project.

1 See Equality Between Women And Men — 2009 Report From The Commission To The European Parliament, The Council, The European Economic And Social Committee And The Committee Of The Regions (Com(2009) 77 Final).

2 Making Equality Effective: The role of proactive measures

Several reasons for this emerge from the reports from the Member States. First, the complaints-led model requires the perpetrator to be identified and proof that the law has been breached. Even with a shift in the burden of proof, it remains very difficult to prove discrimination. Problems of proof are exacerbated in the case of equal pay by the need to find a comparator. This is true too of indirect discrimination, which is a complex concept, raising difficult questions as to appropriate comparisons. A second and particularly significant deterrent is the fear that by initiating legal proceedings, the victim will be exposed or stigmatised as a nuisance. Thirdly, judicial procedures are slow and, in most Member States, costly. Fourthly, in most Member States, remedies are retrospective and the amount of compensation available is not sufficient to deter the perpetrator nor make it worth the victim’s while to pursue a case. In addition, in some Member States, there is a low level of awareness among the public in general of discrimination law, and a lack of familiarity among the legal profession itself. The result is that few victims pursue their claims through court, and many breaches go unremedied. It is striking that this occurs even when legal aid is readily available and court costs are not high; or when remedies are real and dissuasive. This suggests that any permutation of the above reasons is sufficient to deter complainants. Paradoxically, too, even when the complaints-led approach is well utilised, its effects are limited, since courts are unable to cope with a large number of individual cases, exacerbating delays. Finally, and most importantly for our purposes, the complaints-led model only addresses discrimination where there is an identifiable perpetrator. In fact, inequality between men and women has causes which go well beyond individual actions. This means that, in addition to a complaints model, it is important to explore the possibilities of establishing processes which facilitate structural and institutional change. 2. Modified complaints-led models Many jurisdictions have introduced modifications to the pure complaints-led model to mitigate the difficulties faced by the claimant. There are several complementary methods by which modifications have been introduced. One involves lessening the burden on the individual by widening standing rules. A significant number of Member States allow equality bodies, NGOs or trade unions to pursue the case on behalf of the victim, and in some cases, procedures can be initiated even if there is no immediately identifiable victim. Secondly, many jurisdictions have given adjudicative functions to alternative bodies, which are much faster and cheaper than a fully judicial approach. A third approach is to establish alternative dispute procedures which depart from the adversarial model and aim to address complaints of discrimination through compromise and consensus. Modified complaints-led models make an important contribution to addressing the limitations of a pure complaints led approach. This is particularly true where collective complaints can be brought and where forward looking remedies may be triggered by individual or group complaints. Nevertheless, although modified complaints-led models deal with a larger number of complaints than judicial procedures, available statistics show that they are still not widely used. This was explained in various ways in the reports. Many mentioned lack of widespread knowledge of these processes by the victims themselves and by union representatives, lawyers, judges and labour inspectors. Absence of meaningful sanctions is also referred to as a possible cause.


3. Proactive measures More recently, attention has shifted to the possibility of alternative and complementary means of addressing gender inequality. These focus on measures to promote or achieve equality between men and women. Rather than being initiated by individual victims against individual perpetrators, responsibility is placed on bodies, such as public authorities or employers, who are in a position to bring about change, whether or not they have actually caused the problem. Such models aim to remedy each of the deficiencies of the complaints-led model above. Firstly, instead of consisting in reactions to ad hoc claims brought by individuals, the initiative lies with policy makers and implementers, service providers, employers and trade unions. This relieves individual victims of the burden and expense of litigation. Secondly, change is systematic rather than random or ad hoc, ensuring that all those with a right to equality are covered. The institutional and structural causes of inequality can be diagnosed and addressed collectively and institutionally. Thirdly, in recognition of the institutional basis of discrimination, there is no need to prove discrimination or find a named perpetrator. Instead, the duty to bring about change lies with those with the power and capacity to do so. Rather than determining a breach of the law, the focus is on identifying systemic discrimination and creating institutional mechanisms for its elimination. Finally, proactive models broaden the participatory role of civil society, both in norm setting and in norm enforcing. In this sense, the citizen is characterised not as a passive recipient but an active participant. These measures are known by many different names, including mainstreaming, proactive measures, positive action and others. Because of the confusion in terminology, the term ‘proactive measures’ is used in this study to designate the alternatives to a complaints-led model. This is because, under whatever name, the essence of these models is that they are forward-looking, requiring bodies2 to take the initiative rather than merely responding to complaints. The idea of this study is to capture the many types of measures in Member States which are proactive in that they aim to change existing practices, scrutinise future practices for their impact on women, or implement express policies to further equality for women. It aims to investigate the nature and extent of such proactive measures in Member States, and demonstrate the range of possible models, making it possible for Member States to learn from each others’ experiences. The report is not confined to measures which are obligatory. Most measures include an element of discretion, and while some impose legally binding duties, others are based on incentives, political accountability or goodwill. While including a wide range of such measures, the report does not of course cover all possibilities. In particular, it does not address procurement, which is comprehensively dealt with elsewhere.3 3.1 Eliminating unlawful discrimination: the case of equal pay The report examines three different ways in which proactive measures operate. The first is to provide more effective means of ensuring that existing anti-discrimination

2 The term ‘bodies’ is used here to include a wide variety of actors, including public authorities,

Member States, civil servants and other government officials, equality bodies, unions, employers and others who are in a position to bring about change.

3 See C. McCrudden Buying Social Justice: Equality, Government Procurement, and Legal Change Oxford and New York: Oxford University Press 2007; C. McCrudden ‘Advice to a Legislator on Problems Regarding the Enforcement of Anti-Discrimination Law and Strategies to Overcome Them’ in: T. Loenen and P.R. Rodrigues Non-Discrimination Law: Comparative Perspectives pp. 295-312 Kluwer Law International 1999.


laws are fulfilled. Instead of responding only to individuals with the courage and resources to bring a complaint to a court or tribunal, the employer or public body should take the initiative in seeking out instances of unlawful discrimination and rectify them. This means that the right to equality is available to all, not just those who have the courage and resources to bring a complaint to court or tribunal. Moreover, rather than redressing unlawful discrimination only for the benefit of a particular individual, such an approach seeks to find collective solutions, covering all affected individuals. As a case-study of this type of proactive measure, the report examines equal pay, an area in which the weaknesses of the complaints-led model are particularly acute. As we have seen, despite the fact that the right to equal pay for equal work has been part of EU law since the original Treaty of Rome, there remains a significant gender pay gap throughout Europe. Proactive measures might therefore constitute an important complementary strategy for tackling unequal pay for women. Nevertheless, it is striking how many Member States have taken no proactive measures in relation to pay. Among those Member States that have included equal pay among their proactive measures, many halt at the collection of gender-disaggregated statistics. A few states have, however, taken a further step and established mechanisms to identify possible solutions. Possibly most important are those measures which tackle pay inequalities by concentrating on pay structures at enterprise or occupational level, where the primary causes of the gender pay gap are found. In several Member States, employers are required to formulate periodic equality action plans, usually in cooperation with trade unions or other employee representatives. Other Member States give collective bargaining a central role in addressing pay inequality, requiring gender objectives to be integrated into general negotiations on wages. In one Member State, fixed targets have been set for the elimination of the gender pay gap. 3.2 Preventing inequality: impact assessment In addition to eliminating existing unlawful discrimination, proactive measures aim to prevent inequality arising in the first place. A key means to achieve this is to require decision makers to assess new policy or legislative measures to determine their impact on gender and to adjust them if necessary. While a significant number of Member States have not instituted such measures, impact assessment is receiving increasing attention in several Member States. This is particularly true at legislative and executive level, where a significant number of Member States have instituted impact assessment of draft legislation and policy. In a few cases, impact assessment is also carried out by equality bodies or even private employers. However, although a number of Member States carry out the initial assessment, few ensure that further action is taken once a negative impact has been identified, and even fewer review changes once instituted. The more sophisticated provisions for impact assessment are relatively recent, and the efficacy of the duty to assess the impact of previous planning is yet to be seen. Impact assessment might also be impeded by the shortage of resources. In particular, where responsibility lies with the equality body, appropriate resources are necessary if their task is to be comprehensively fulfilled. 3.3 Promoting equality: structural change A third function of proactive measures is to look beyond existing anti-discrimination law and seek to promote equality between women and men regardless of whether there is evidence that a particular individual has been guilty of discrimination. Such an approach is capable of addressing structural discrimination where no specific


individuals are responsible. The study examines two methods of promoting gender equality: quotas and family-friendly measures. The use of quotas or set-asides to promote gender equality is inevitably controversial, and this mechanism remains unlawful in several Member States. Nevertheless, in a number of Member States, quotas are regarded as a means of achieving equality, rather than a breach. In these Member States, quotas have been effectively used in a number of contexts. Quotas for women in the legislature, whether local, national or European, are more common than might be expected and in several countries, political parties have voluntarily adopted quotas. A handful of Member States have also instituted quotas in decision-making bodies and in public employment. On the other hand, very few countries have quotas in respect of the judiciary or the private sector. Several reports note that quota provisions have been the most effective form of proactive action insofar as gender is concerned. This does, however, require effective sanctions and depends to a large extent on the political culture. The second method of promoting equality dealt with in the study relates to family-friendly measures. One of the key causes of gender inequality is the fact that women remain primarily responsible for child-care and housework. To be effective, it is therefore essential to include proactive measures which both facilitate women’s involvement in the paid workforce, and enhance men’s ability to participate in child-care responsibilities. Almost all Member States include family-friendly measures in their constellation of proactive measures. However, the most striking finding was that the focus remains primarily with the mother. Provision of leave for fathers is considerably shorter than for mothers and is frequently unpaid. Even when leave or rights to reduced working time are optional and available to both mothers and fathers, the uptake by fathers remains low, a pattern which is consistent across the vast majority of Member States. This remains true too when the right to leave cannot be transferred from the father to the mother and is therefore lost if not taken. A primary reason is the fact that parental leave is unpaid, or, if paid, attracts compensation at too low a level to create sufficient incentives for fathers. Added to the low level of payment for fathers are cultural expectations, which militate against fathers staying at home to look after children. Experience from some Member States shows that fathers’ uptake of paternity and parental leave can be significantly improved by increasing the remuneration available to men. Moreover, changing the gender stereotypes as to the division of care responsibilities between men and women is seen as an important measure in achieving equal opportunities for men and women. 4. Ingredients of a proactive model 4.1 Responsibility The study examines four central ingredients of a proactive approach: responsibility, participation, monitoring and enforcement. Proactive measures entail a shift in responsibility away from the individual claimant to a body which is in a position to take action to eliminate unlawful discrimination, to assess new policies for their impact on gender equality and to promote structural equality.. The question of who has such responsibility is therefore of central importance. Public bodies have responsibility for taking proactive measures in 13 Member States; and in a significant number of Member States, responsibility for proactive measures also falls on trade unions and public or private employers. More specific allocation of responsibility is more effective at delineating responsibility than vague or general provisions, which run the risk that no specific person takes on the responsibility. The provision of a


network of institutions, as well as channelling responsibility through collective bargaining procedures, tend to be the most effective at pinpointing responsibility. However, the risk remains that the apparatus of responsibility exists, but little action is taken in practice. As in other respects, much depends on political commitment and goodwill. 4.2 Participation A second element is participation. Given the potential bureaucratic and ‘top-down’ nature of proactive measures, it is of central importance to involve stakeholders, potential victims, trade unions, service users, relevant NGOs, and others in the process. Consultation of some sort is found in the vast majority of Member States. Particularly evident is the growing trend towards canvassing the views of stakeholders on draft legislation or policy measures. The study also examines the method chosen to identify consultees: in the absence of a specified victim, questions arise as to the representativeness of consultees, their expertise, and their capacity to engage in the process. However, there is typically no systematic principle for selection of consultees. Several Member States have created a durable consultative framework, involving equality bodies, members of the Government, social partners and other interested parties, while others give this role primarily to the equality body. Collective bargaining or works council structures also constitute an important arena for consultation at the enterprise level. Not all consultative fora, however, have been functional or effective. Also of importance is the function of consultation. Consultation is invariably aimed at giving and gaining information: no Member States reported consultation with binding effect on decision making. In some Member States, decision makers are simply required to consider the views of consultees; in others, rejection of such views must be accompanied by reasons. Several Member States delegate the drafting process itself to the consultees (usually the equality body). It is clear that, rather than having binding force, the weight given to opinions of consultees depends largely on the political culture, the goodwill of decision makers and the political or industrial strength and influence of consultees. An active and engaged civil society is therefore crucial to the success of participation mechanisms. 4.3 Monitoring Monitoring is a third essential ingredient of a proactive approach. Unlike an individual complaints model, which is concerned with a self-contained incident, proactive measures are programmatic and on-going. A process of monitoring and review is therefore essential to assess whether a proactive measure is effective, to review its progress, and to readjust it if necessary. As a first step, it is necessary to gauge the extent of the problem, usually through constructing a statistical picture. While a number of Member States still do not collect gender-disaggregated statistics at the national level, in other Member States, there are now quite comprehensive duties to collect such statistics. Alternatively, monitoring may be conducted through requiring regular reports from duty-bearers, to be assessed by a monitoring body such as the equality body or worker representatives at enterprise level. However, it is not sufficient simply to collect data or reports. It is also important to make use of these instruments to assess and review the progress of proactive measures. In some Member States, this link is not made. Several Member States have, however, instituted sophisticated structures for reviewing proactive measures in the light of the outcomes of monitoring, and taking of remedial steps. In practice, however, it is not clear that


monitoring obligations are carried out as well as they should be. One possible obstacle to monitoring might be that the collection of statistics might breach domestic confidentiality or data protection laws. In general, there are no such obstacles as long as individuals are not identified. However, a key source of difficulty arises in respect of equal pay, where in many Member States wages are regarded as confidential. This makes it close to impossible to identify and monitor gender wage gaps at the enterprise level. A handful of Member States have dealt with this difficulty by permitting workers to disclose their wages. 4.4 Enforcement and compliance The final section of the report examines enforcement and compliance. Without the ultimate sanction of judicial procedures, there is a risk that proactive measures might become mere rhetorical gestures. A key challenge for proactive measures, therefore, is to devise appropriate means of enforcement. This has proved to be the most problematic aspect of proactive duties. Much depends on political goodwill and a sense of responsibility on the part of duty-bearers, and when these are lacking, there is no easy solution. Several possible enforcement mechanisms are canvassed. One is to require regular reporting to Parliament, thereby relying on the political process to ensure that proactive measures are carried out. The practice of reporting to Parliament is widespread among Member States. The effectiveness of the reporting mechanism is, however, dependent on the seriousness with which it is regarded by Parliament or the relevant Ministry and whether further action is taken. There is little information on how this process works in most jurisdictions, although some referred to the risk that the reporting procedure might become no more than a formality. A second possibility is to give enforcement powers to equality bodies or other similar institutions. Such powers are given to the equality body in eleven Member States. The ideal model in this context would be a pyramid of enforcement, whereby the first response to non-compliance would be for the equality body to initiate a process of discussion and negotiation. If this is not successful, the recalcitrant respondent could be subject to an order to comply issued by the equality body. Only if this further step fails do fines or other judicially enforced sanctions come into play. A handful of Member States follow a permutation of this pattern. A third possibility is to enforce legal obligations through collective bargaining structures. In countries without a well-developed collective bargaining structure, this means of enforcing proactive measures is conspicuously absent. Several reports point to the need to retain some initiative for individuals within a proactive model. Without reverting to an individual complaints model, pro-active measures might nevertheless better serve the objective of achieving real and substantive equality if tools were granted to the victims themselves. Most Member States have no such possibility, although a handful of Member States do allow individuals to bring complaints where proactive measures have not been fulfilled. Judicial review is particularly rare, applying only in a small number of Member States. More promising is the potential for trade unions to complain in these circumstances, a possibility in both Sweden and France. There is little experience of the workings of enforcement powers in the context of proactive measures. However, several difficulties are revealed in this study. The first is that proactive measures are themselves ill-defined and therefore it is not easy to determine whether they have been breached. This issue is identified in several reports, where it is pointed out that sophisticated enforcement powers are undermined by the


absence of concrete measurable objectives designed to assess the extent to which proactive measures have been implemented. A second difficulty concerns the lack of resources allocated to the fulfilment of proactive measures, as well as to monitoring and enforcement. This problem is exacerbated by the current recession, which is drawing resources away from this area. The result is that the apparent development from formal to substantive equality, and from complaints-led models to proactive measures, may not truly reflect the real situation. 5. Conclusion The introduction of proactive models into the arena of gender discrimination holds much promise, as well as many new challenges. The architecture of proactive measures is increasingly visible within Member States’ provisions for gender equality, and there is a continuing and dynamic process of development and appraisal. However, structures on their own are not sufficient to ensure that proactive measures are implemented and have an impact. The greatest challenge remains that of creating appropriate incentives, sanctions and mechanisms for accountability to ensure that elaborate structures do not simply conceal apathy or proceduralism. The study suggests that political commitment and good will, together with the active involvement of stakeholders, particularly trade unions and other representatives of those affected, are essential for success. The danger remains that the location of proactive measures on the borderline between law and politics makes it appear that fulfilment of such measures is discretionary or optional. The ultimate challenge is therefore to ensure that proactive measures are based on a recognition that equality is a fundamental right, not an optional policy. Moreover, proactive measures and the complaints-led model should not be regarded as mutually exclusive. They both have their own peculiar function. The complaints-led model, where successful, makes it possible to satisfy the specific needs of the individual claimant; whereas proactive measures are normally more sensitive to the general needs of the victims of the discrimination. Particularly interesting is the possibility of a creative synthesis between a judicial and a proactive model. This can be created by giving powers to judges or equality bodies to order collective and forward-looking solutions. In Italy, for example, where an Equality Adviser brings a case of collective discrimination to court, the judge can issue an order to the guilty party to carry out a plan to remove the discrimination. Equality Advisers can also propose a conciliation agreement before going to court, asking the person responsible for the discrimination to set a plan to remove it within 120 days. If the plan is considered fit to remove discrimination, on the Equality Adviser’s demand, the parties sign an agreement which becomes a writ of execution through a decree of the judge. Similarly, in Ireland, a successful claimant can obtain significant redress and at the same time there may be an order of the Equality Tribunal or Labour Court that a course of action should be adopted, for example to change a recruitment process.


I INTRODUCTION Gender inequality across Europe remains elusive. Female employment has increased consistently from 54.4 % to 58.3 % between 2002 and 2007. However, this is not in itself evidence of a reduction in inequality. Figures from 2007 show that across Europe women earn on average 17.4 % less than men. The pay gap4 exceeds 25 % in two countries5 and 20 % in seven countries. 6 In only five countries7 is it is below 10 %.8 Labour markets continue to be highly segregated and women still cluster in women’s only jobs, or predominate in the low levels of mixed jobs.9 Occupational segregation across the EU as a whole has remained relatively high, with little change since 1992 apart from a slight upward trend over the current decade. Indeed, there is evidence that significant increases in female employment may raise the level of segregation on a more or less temporary basis.10 Moreover, the increased participation of women in the labour market is largely characterised by a high proportion of part-time work. In 2007, the percentage of women employees working part-time was 31.2 % in the EU-27 while the corresponding figure for men was 7.7 %. The share of female part-timers exceeded 30 % in France, Ireland, Denmark and Luxembourg and 40 % in Sweden, Belgium, Austria, United Kingdom and Germany and even reached 75 % in the Netherlands.11 This reflects the fact that many women bear a double burden: they remain primarily responsible for child-care, care of the elderly and the disabled while at the same time undertaking paid work. Working women spend, on average, more time than working men in domestic and family work in all Member States and have on average more ‘constrained time’ (sum of hours spent in paid work and in unpaid domestic and family work) than working men.12 Yet so far as the provision of child-care facilities is concerned, in 2006, only five Member States13 had exceeded the Barcelona objective of a 33% coverage rate.14 Figures show that parenthood has a significant long-term effect on women’s participation in the labour market, with a significant decrease in women’s employment rate when they have children under 12, compared to a significantly higher employment rate of men with children under 12.15

4 Defined as the difference between men’s and women’s average gross hourly earnings as a

percentage of men’s average gross hourly earnings. 5 Estonia and Austria. 6 SK, NL, CZ, CY, DE, UK and EL. 7 Italy, MT, PL, SI and BE. 8 See Equality Between Women And Men — 2009 Report From The Commission To The European

Parliament, The Council, The European Economic And Social Committee And The Committee Of The Regions (Com(2009) 77 Final).

9 For the EU as a whole, the level of segregation, as measured by the IP index, is still relatively high, with a 25.3 % value for occupational segregation and a value of 18.3 % for sectoral segregation out of a maximum of 50 %.

10 F. Bettio and A. Verashchagina Gender segregation in the labour market: Root causes, implications and policy responses in the EU European Commission’s Expert Group on Gender and Employment (EGGE) European Commission Directorate-General for Employment, Social Affairs and Equal Opportunities (March 2009).

11 Equality Between Women And Men — 2009. 12 Ibid. 13 DK, NL, SE, BE, ES: ibid 14 Ibid. 15 In 2007, the employment rate for women aged 25-49 was 65.5% when they had children under 12,

compared with 77.9% when they did not, a negative difference of 12.4 percentage points (p.p) Interestingly, men with children under 12 had a significantly higher employment rate than those


Inequality is not confined to the labour market. The average number of female members of national parliaments (single/lower houses) was 24 % in 2008, one percentage point higher than in 2004. The percentage was above 35 % in Spain, Belgium, Denmark and the Netherlands and above 40 % in Finland and Sweden. However, it was below 15 % in Ireland, Slovenia and Hungary and did not exceed 10 % in Malta and Romania.16 The tenacity of gender inequality suggests the need to re-examine the methods used to achieve real and substantive equality between men and women. The traditional approach has been to rely on an individual complaints model of adjudication. But a range of new approaches are emerging, which aim at institutional change through proactive measures to promote equality. This report assesses these models in the light of current practice among the Member States of the EU and EEA. For the purposes of this report, experts from the 27 EU Member States and Iceland, Liechtenstein and Norway filled in the questionnaire at Appendix 1. The complaints-led model, when widely utilised, can be an important source of remedies for an individual victim. It might also function as a deterrent to potential perpetrators, and trigger changes in legislation and policy. However, it is limited in at least three ways. Firstly, reliance on an individual complainant to bring an action in court puts excessive strain on the victim both in terms of resources and personal energy. Litigation is lengthy and costly. Secondly, victim initiated litigation means that the court’s intervention is random and ad hoc. Many individuals, particularly non-unionised ones, are unable to pursue their claim. The result is that a large number of cases of discrimination go unremedied. Thirdly, it is necessary to identify a perpetrator and prove a breach of anti-discrimination law. Yet it is now recognised that much inequality is institutional and not the fault of any one person. These limitations make it difficult to initiate and substantiate claims, undermining the potential of increasingly sophisticated anti-discrimination laws, such as prohibitions of harassment, pregnancy discrimination, and indirect discrimination. More recently, attention has shifted to the possibility of alternative and complementary means of addressing gender inequality. Some of these widen the complaints-led model to permit collective complaints. Others go further and place the initiative on public bodies, employers, and others to tackle discrimination and promote equality even in the absence of a complaint. Such models aim to remedy each of the deficiencies of the complaints-led model above. Firstly, instead of consisting in reactions to ad hoc claims brought by individuals, the initiative lies with policy makers and implementers, service providers or employers. This relieves individual victims of the burden and expense of litigation. Secondly, change is systematic rather than random or ad hoc, ensuring that all those with a right to equality are covered. The institutional and structural causes of inequality can be diagnosed and addressed collectively and institutionally. Thirdly, in recognition of the institutional basis of discrimination, there is no need to prove discrimination or find a named perpetrator. Instead, the duty to bring about change lies with those with the power and capacity to do so. Instead of determining breach by an identified perpetrator, the focus is on systemic discrimination and the creation of institutional mechanisms for its elimination. This means that the right to equality is potentially available to all, not just those who complain. Finally, proactive models broaden the participatory role of civil

without, 91.7% compared with 84.4%, a positive difference of 7.3 p.p: Equality between men and women.

16 Ibid.


society, both in norm setting and in norm enforcing. In this sense, the citizen is characterised not as a passive recipient but an active participant. The EU itself has given particular priority to such measures, both in Article 2, which highlights equality between men and women as a common value of the EU, and Article 3(2), which specifically states that the Union shall promote equality between women and men. A variety of such models has been introduced both in Member States and in the EU itself. These measures are known by many different names, including mainstreaming, proactive measures, positive action and others. Because of the confusion in terminology, the term ‘proactive measures’ is used in this study to designate the alternatives to a complaints-led model. This is because, under whatever name, the essence of these models is that they are forward-looking, requiring bodies17 to take the initiative rather than merely responding to complaints. The idea of this study is to capture the many types of measures in Member States which are proactive in that they aim to change existing practices, scrutinise future practices for their impact on women, or implement express policies to further equality for women. It aims to investigate the nature and extent of such proactive measures in Member States, and demonstrate the range of possible models, making it possible for Member States to learn from each others’ experiences. The report is not confined to measures which are obligatory. Most measures include an element of discretion, and while some impose legally binding duties, others are based on incentives, political accountability or goodwill. While including a wide range of such measures, the report does not of course cover all possibilities. In particular, it does not address procurement, which is comprehensively dealt with elsewhere.18 II THE COMPLAINTS-LED MODEL All the Member States surveyed make it possible for victims of discrimination to invoke a judicial procedure, as required by EU law. 19 It was pointed out above that such procedures are in principle limited: they leave the initiative to the individual victim, and only address discrimination where there is an identifiable perpetrator who can be proved to have breached the law. This means that the impact of equality is likely to be patchy: structural discrimination cannot be addressed if there is no identifiable perpetrator, and cases of unlawful discrimination go unremedied if no victim is willing and able to come forward. This section examines the way in which the complaints-led model operates in the Member States. It is difficult to gauge precisely whether the complaints-led model is sufficiently widely used to make a real impact on discrimination. This is because of the striking fact that very few Member States compile statistics on the use of courts in discrimination cases. No statistics were available in respect of court proceedings in Belgium, Denmark, Finland, Germany, Greece, Hungary, Ireland (in respect of the

17 The term ‘bodies’ is used here to include a wide variety of actors, including public authorities,


18 See C. McCrudden Buying Social Justice: Equality, Government Procurement, and Legal Change Oxford and New York: Oxford University Press 2007; C. McCrudden ‘Advice to a Legislator on Problems Regarding the Enforcement of Anti-Discrimination Law and Strategies to Overcome Them’ in: T. Loenen and P.R. Rodrigues Non-Discrimination Law: Comparative Perspectives pp. 295-312 Kluwer Law International 1999.

19 EP and Council Directive 2006/54, Article 17(1).


Circuit Court), Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, the Netherlands, Portugal, Slovenia and the UK (for county courts). Nevertheless, in the few states in which such statistics were available, it was clear that only a handful of gender discrimination cases are litigated before courts each year. By 2008 in Slovakia, there had been no court decisions since the Anti-discrimination Act came into force in 2004. Under the previous jurisdiction, the anti-discrimination provision in the Labour Code, there were only three successful cases, one in the court of first instance (the District Court) and two before the Supreme Court, the latter both being against the same defendant. In Bulgaria, 2 court decisions on gender discrimination were reported in 2006–7 of which only one was successful. In Cyprus, there were 3 gender discrimination cases before the Supreme Constitutional Court in 2002–3. The Industrial Tribunal Court, which has jurisdiction over employment discrimination and equal pay issues, dealt with only 1 case in 2005 and one more in 2008. Although this court has power to order the re-employment of the worker, there have been no cases on this issue thus far. A similar pattern is evident in other Member States. In the Czech Republic, the only available statistics show that there was one case of discrimination in employment in 2005, 5 in 2006 and 3 in 2007; while on the pay gap there were 5 in 2006. In Estonia, the first case on gender equality to reach the Supreme Court was being heard in 2009. In Iceland, the Supreme Court delivered only 7 judgments concerning alleged discrimination in breach of the Gender Equality Act between 2000 and 2009 and acquitted the Icelandic state in two of them.20 In France, although the Cour de cassation issued 111 decisions on discrimination in 2007 and 139 in 2006, very few were specifically concerned with sex discrimination, with the majority concerned with discrimination on grounds of trade union membership. Where statistics were not available, expert reporters had a strong impression that litigation in this area was rare. For example, in Belgium, it was reported that there are fewer and fewer reported cases of gender-related discrimination, with no more than 5 during the last year. In Liechtenstein, the remedial procedures have been used rarely or not at all. Similarly, in Portugal, although no statistics are available, judicial complaints regarding discriminatory practices are rare, while in Sweden, very few complaints are brought to court. Several reasons consistently emerge for the paucity of litigation through the pure complaints-led model. The first is that such procedures are slow and costly. For example, the most important problem identified in respect of the complaints-led model in the Czech Republic is the fact that the procedure often lasts for many years and is very expensive. This low level of trust in the judicial system is reflected too in Romania, where the general perception is that seeking justice is too expensive, time consuming and complicated, particularly because of long delays. Long delays were also identified as problematic in Slovenia while the Latvian report pointed to the anxiety induced by court procedures. Problems of proof constitute a second disincentive. In Romania, for example there is a widespread perception that gender-based discrimination is very difficult to prove, especially in the labour field, and that it is futile to try to use the judicial system to seek justice. Similarly, in Slovenia, despite the shift in the burden of proof, discrimination is very difficult to prove, especially in cases of indirect discrimination or harassment. In Greece, the burden of proof rule is inadequately transposed as regards its scope, in that it applies to courts only, not to other competent authorities.

20 Supreme Court cases no. 121/2000 and no. 350/2004.


Nor is the burden of proof included in the procedural codes; as the Council of State recommended.21 It is thus not known and not applied. Moreover, in 14 jurisdictions, it is necessary to identify a victim before a claim can be brought. As well as precluding the possibility of addressing structural discrimination, this leaves the initiative entirely in the hands of the individual complainant, a burden which is exacerbated by the absence of standing for trade unions, NGOs and other bodies to bring claims. The result is that even sophisticated legislation addressing gender discrimination can make little impact on patterns of discrimination. As the Portuguese report shows, generally good and updated legislation subsists alongside bad practice and little jurisprudence with very few cases brought before the courts, due to economic and psychological deterrents. Thirdly, victims may be discouraged by the low quality of court judgements. This is true of the few decisions already taken by the Czech courts. This may be accompanied by lack of familiarity among the legal profession itself, as in Slovenia, where the judiciary itself is not accustomed to gender discrimination disputes and lacks familiarity with the application of EU law. A more general problem is the low level of awareness on the substance of discrimination and the available legal remedies, a problem strongly emphasised by the Lithuanian and Slovenian reports. A fourth significant deterrent is the fear that attempted proceedings might make the victim feel exposed or stigmatise her as a nuisance. This has been particularly problematic in Iceland for individual civil servants who wish to claim discrimination, after a Supreme Court ruling overturned both the opinion of the Complaints Committee and the judgement of the district court by holding that there had been no violation of the Gender Equality Act when the foreign minister appointed a man for a high-ranking post and not a woman.22 Similarly, in a small country like Liechtenstein, where employers know each other very well, women might feel that it is not sensible or desirable to choose a court procedure to solve a discrimination case. A further reason – identified with respect to Latvia – is the lack of visibility of discrimination cases even when they do reach the courts. This lack of public debate in the mass media and elsewhere means that even when cases are decided on an individual matter, they cannot contribute effectively to the changing of stereotypes and attitudes in general. It is striking that levels of litigation remain very low even when there is a well-organised free legal aid system for those who cannot afford a lawyer and even when legal costs are not too high, as in Slovenia. Nor is the situation necessarily improved by effective sanctions. Thus in Greece, levels of litigation remain low, despite the relatively powerful remedies and sanctions available. For example, a discriminatory dismissal may be declared null and void (by civil courts) or be annulled (by administrative courts). Since the dismissal is deemed never to have occurred, the worker retains her post, and thus, reinstatement is unnecessary. Furthermore, a discriminatory refusal to hire or promote can be declared null and void by the civil courts so that the hiring or promotion is deemed to exist from the time it should have occurred. Correspondingly, administrative courts may annul such a refusal and order the issuance of an administrative act of retroactive hiring or promotion, awarding the worker full back pay plus legal interest and, where relevant, compensation for future financial loss and moral damages. Heavier sanctions are provided for a ‘violation of sexual dignity’ (a criminal act under the Penal Code), when it constitutes the 21 Council of State (Supreme Administrative Court) Opinion 348/2003 on the draft decree transposing

the burden of proof directive. 22 No. 121/2003.


exploitation of a worker or candidate for work. Nevertheless, even in Greece, aggrieved women are reluctant to claim their rights. Lack of evidence, fear of victimisation and of acquiring a ‘bad name’ in the labour market (unemployment is higher for women than men) weigh more heavily on women than the promise offered by effective sanctions. Also an important reason for low litigation levels is the inadequate transposition and non application in practice of EC procedural rules, particularly the rules on burden of proof, and on the standing of organisations to pursue claims on behalf of victims of discrimination,. There is one notable exception to the pattern of underutilisation of the pure complaints-led model. Statistics from Poland show a steady increase in the use by women of the available court procedures to pursue gender discrimination claims:23 – In district courts, in 2002 (the first year in which gender equal treatment

provisions were binding), there was only 1 such a claim. – In 2003 this had grown to 20 (of which 15 were filed by women). – In 2004, this had jumped to 238 (of which 128 filed by women). – In 2005, there were 220 claims (139 filed by women). – In 2006, 195 (95 filed by women), and in 2007, 219 (190 filed by women). – By 2009, there were as many as 989 claims of which 698 were filed by women. This is particularly striking in that Poland has not substantially modified the adjudicative model; the equality bodies do not have powers to bring claims on behalf of individuals to court or to conduct investigations although they do have power to give advice. On the other hand, district courts are reasonably easy to access. Until 2006, proceedings related to employment were totally exempt from court fees. Although this principle was abolished in 2006, court fees are not high.24 This is contrasted with circular courts (hearing more serious cases). – In 2003 there was only 1 claim (filed by a man). – In 2004, there were 36 claims (17 filed by women). – In 2005, there were 46 claims (23 filed by women). – In 2006, there were 34 claims (14 filed by women). – In 2007, there were 36 claims (16 filed by women). – In 2008, there were 39 claims (20 filed by women). Nevertheless, the numbers of successful court claims, as well as the percentage of total cases completed within one year are relatively low.25 – In district courts only two out of 13 completed cases (15 %) were successful in

2003, and 3 claims were unsuccessful (23.1 %).26 – In 2004, out of 133 completed claims, 2 were successful (1.5 %) and 11

unsuccessful (8.2 %).27 23 While analysing the numbers of claims in the indicated year it should be realised that it is composed

not only of the new claims but also of undecided cases from preceding years. 24 The situation becomes different if the amount of litigated compensation exceeds 50 000 PLN (circa

EUR 11 600), because then the general court fees rules in civil cases apply (fee is equal to 5 % of value of object of claim).

25 It should be mentioned that the number of ‘finished cases’ also embraces cases closed in another way than positive or negative court decisions (e.g. the claim has been transmitted to another organ or cumulated with another claim).

26 In one case an agreement was reached (7.7 %).


– The success rate in the next years remained low and was approximately 2 to 4 times lower than failure rate. In 2005, 21 claims were successfully closed, of 145 completed cases (14.5 %) while the failure rate was 24.8 %.

– In 2006 of 132 finished claims, 14 were successful (10.6 %) and 57 unsuccessful (43.2 %).

– In 2007, of 209 completed cases, 28 claims were successful and 47 unsuccessful (13.4 % and 22.5 %).

– In 2008, of 824 recognized claims, 47 were successful and 75 unsuccessful (5.7 % and 9.1 %).

– There was a steady proportion of cases settled by agreement in the period 2005-8 (the highest being 13 in 2007 constituting 6.4 % of all cases completed that year).

It is also worth comparing the success rate of men and women. – In the years 2005 and 2006, the success rate in case of claims lodged by women

did not substantially differ from that of men. In 2005 there was a 14.9 % success rate in case of women and 13.7 %, in case of men. In 2006, the rate was respectively 11.3 % and 10 %.

– However in the years 2007 and 2008, success rates for women were significantly lower than for men (respectively 7.1 % to 20 % and 4.1 % to 9.8 %).

The success rate in circular courts was also low, even in comparison with district courts. – In 2004, there were no successful cases. – In 2005, one case was successfully closed and 14 claims were unsuccessful,

constituting 3.6 % and 50 % of all examined claims. – In 2006, out of 23 completed cases, 5 were successful and 10 unsuccessful

(21.7 % and 43.5 %). – In 2007, out of 14 completed cases, none were successful and 5 were unsuccessful

(36.7 %). – In 2008, out of 15 completed cases, 2 were successful and 5 unsuccessful (13.3 %

and 33.3 %). Nor are sanctions particularly dissuasive. Although there are no reliable data as to the level of monetary compensation awarded in cases of gender discrimination, on the basis of some court decisions it is estimated that the average level of compensation awarded where discrimination is found varies from 1 to 10 times28 the minimum monthly wage (from PLZ 1 276 to 12 276, circa EUR 300 - 2 854). Compensation of this amount is likely to have little dissuasive effect. III MODIFIED COMPLAINTS MODEL Many Member States, recognising the difficulties with the individual complaints model, have modified it in important ways. Others have not expanded beyond the pure complaints model. An example is the Czech Republic, which has introduced none of the modifications found in other jurisdictions. Malta is in a similar position. 27 In 2 cases an agreement was reached (1.5 %). 28 See e.g. Supreme Court decision of 7 January 2009 (III PK 43/08) www.sn.pl/orzecznictwo/.html,

accessed 28 August 2009.


There are two main, complementary methods by which modifications can be introduced. The first involves lessening the burden on the individual by allowing actions to be initiated by others. The second consists of the provision of alternative forms of adjudication which are cheaper and more accessible. 1. Minimising the burden on the individual One way of lightening the burden on the individual is to allow class actions, whereby a group of victims can litigate the same claim together. About half the Member States (11) permit class actions (Bulgaria, Cyprus, Denmark, Italy, Lithuania, the Netherlands, Norway, Portugal, Slovakia, Spain and Sweden). In Greece, class actions may be lodged by consumers’ associations for breaches of consumer protection legislation. In principle, class actions should therefore be permitted for breaches of the statute transposing Directive 2004/113/EC, the more so as this statute designates the Consumer’s Ombudsman as the equality body required by this Directive. However, as this statute is very recent, no relevant case law seems to exist. Class actions are, however, only useful in specific circumstances, and where all the victims have an identical claim. By way of contrast, in Hungary, class action is not available in cases where there are identifiable victims. Instead, the subject of any so-called class action must affect a non-identifiable major group of persons. This position is the result of long-standing resistance against the idea of class action in Hungary, and now has a restraining impact on the opportunities for litigation. A different way of taking the burden off the individual litigant is to give standing to other bodies to pursue her case for her. EU law in fact requires that Member States should ensure that standing be accorded to relevant bodies to engage in any judicial procedure on behalf of or in support of the complainant, with his or her approval.29 In Italy, for example, Equality Advisers can bring a claim on the victim’s behalf. This strengthens the victim’s financial and psychological position and ensures the assistance of an expert both before and after the trial. In twelve Member States, equality bodies have the right to bring claims on behalf of individuals (Austria, Belgium, Bulgaria, Finland, France, Ireland, Italy, Latvia, Malta, Romania, Slovakia and Sweden). Notably, this generally requires the consent of the victim. Thus, in Malta, the National Commission for the Promotion of Equality may bring a suit on behalf of the injured party where it has been specifically requested to assist in that regard by the individual.30 The necessary consent on the part of the individual may be withdrawn at any time, resulting in the withdrawal of the case. The NCPE has no capacity to continue on the basis of evidence given should such consent be withdrawn. In Latvia, the Ombudsman may act as the victim’s representative; but the claimant must be the victim herself. Similarly, in Finland, the Equality Ombud may assist a victim in cases of general interest, but in the capacity of legal representative, thus requiring the victim’s consent. In Greece, the statute transposing Directive 2002/73/EC allows a wider circle of organisations to intervene in favour of the claimant, but it does not allow them to bring claims in their own right, thus disregarding the scope of the Directive’s relevant provisions. However, as this rule is not included in the procedural codes, it is unknown, and therefore not applied. A further step would be to permit an equality body to bring a case without an identified victim. This would engage patterns or practices of discrimination on a more

29 EP and Council Directive 2006/54, Article 17(2). 30 Equality for Men and Women Act; Chapter 456 of the Laws of Malta (hereinafter, ‘EMWA’);

Article 12 (1)(j) read in combination with article 11 of the Equal Treatment in Employment Regulations, Legal Notice 461 of 2004 as amended.


collective level, without having to identify a particular perpetrator or a particular victim. There are very few jurisdictions which give standing to an equality body to bring proceedings on their own account – only Austria, Hungary, Italy, Slovakia and the UK fall into this category. In fact, Hungary forbids the bringing of public interest litigation by the equality body or other NGOs when a concrete litigant can be identified. Such public interest litigation is only permitted where no victims are identifiable. In Italy, equality advisers can bring claims of collective discrimination. Also effective, in the employment context, is to permit trade unions to bring such claims. Trade unions can take claims on behalf of victims in 14 Member States (Austria, Belgium, Bulgaria, France, Greece, Hungary, Ireland (only in a representative capacity), Italy, Latvia, Liechtenstein, Lithuania, Romania, Spain, and Sweden). This is particularly useful in countries with high levels of trade union representation. Thus in Sweden, where the rate of affiliation is about 80 %, many victims of alleged discrimination have the advantage of being represented by their trade union, averting some of the costs and stress of litigation. Similarly, trade unionism is very strong in Belgium, where up to 60 % of the potential workforce is organised. However, this does not always take the burden wholly off the individual. Thus in Sweden, a victim must come forward with her complaint even if she is represented by her trade union. Similarly, in Finland, trade unions may provide the victim with a legal representative, but only with the consent of the person concerned. In Lithuania, trade unions have a right to ‘participate’ in court proceedings. However, the ‘participation’ is limited to the procedural representation of the victim with her/his consent or the involvement as a ‘third party without separate claims’. Similarly, in Latvia, trade unions may represent victims, but thus far national courts have not recognised trade unions as having the right to make collective claims. A different limitation is found in Hungary, where the subject matter of the proceedings that can be brought by trade unions is limited to ‘matters related to the living and working conditions or the material, social and cultural situation of workers’. French law takes this one step further and permits a trade union to act on the behalf of an employee claiming to be the victim of discrimination, without having to have a mandate to do so. The only condition is that the employee must be given written notification of the action and has not opposed the union action by the end of a 15-day period. The effectiveness of this approach has recently been demonstrated in the context of discrimination on grounds of trade union membership, where challenges against such discrimination in the enterprise have led to a number of cases in the Cour de Cassation. In Portugal, until 2003, trade unions had a particularly useful role in that they could bring an action before the courts in respect of systematic discrimination. These actions were accompanied by a reversal burden of proof rule, in the sense that it was sufficient that the union proved that the employer acted differently from other economic agents in the same area to give ground to the allegation of discrimination. For instance, a wide difference in the rate of pay between female and male employees in one particular company compared to many other companies of the same sector, would indicate gender discrimination in that particular company. Unfortunately, the Labour Code of 2003 removed this possibility. In Greece, the Code of Civil Procedure (Article 669) allows trade unions to bring claims of their members, in their own name, albeit only if such claims arise from a collective agreement, provided the member does not explicitly disagree. Thus Greek unions may lodge, in their own name, a petition for the annulment of a discriminatory clause of an administrative act of general applicability (acte règlementaire). This includes a decree or ministerial decision, and a collective agreement extended by


ministerial decision. The annulment has retroactive effect erga omnes; the remaining clauses apply ab initio to all those covered by the act. Thus, the annulment amounts to a retroactive modification of the act, and therefore benefits all those covered by the annulled clause. However, the Council of State (Supreme Administrative Court) only admits petitions for the annulment of extended collective agreements when they are lodged by professional organisations which took part in the relevant collective bargaining, not by individual workers covered by the agreement.31 The Greek expert points out that this constitutes a violation of Article 6(1) ECHR32 and the EU general principle of judicial protection. Greek unions may also intervene in favour of a member in any case the latter has initiated. On the other hand, reliance on trade unions can have disadvantages. As the Belgian expert points out, trade unions may decide not to pursue actions, because they do not take the issue au sérieux, or because they prefer a strategy of negotiations with employers rather than a campaign of litigation. Victims of discrimination may find themselves extremely isolated if their unions are reluctant to pick up individual cases or if they are not unionised in the first place. Notably, in Greece, trade unions seldom make use of their rights to bring claims. As the French expert points out, the complaints-led model could be reinforced if the interest of trade unions and workers’ representatives in gender equality could be developed. This suggests too that the most effective approach might be to have a menu of alternative options for an individual complainant, as in Hungary, including trade unions, NGOs, the equality body and others. 2. Modifying the adjudicative body A different method of modifying the pure complaints led model is to give, adjudicative functions to alternative bodies which are faster and cheaper than courts. A third approach is to establish alternative dispute procedures which depart from the adversarial model and aim to address complaints of discrimination through compromise and consensus. A number of member states have given adjudicative functions to alternative bodies. There is a spectrum of possibilities. Closest to the pure judicial model is one which remains complaints based, but reduces the costs, delay, technicality and complexity of judicial procedures by using tribunals dedicated to equality issues or to labour matters more generally. The strongest version of such an approach is a tribunal or board which has power to impose binding sanctions. The tribunal might have its own enforcement powers, as in the UK, or as in the case of Equality Tribunals in Ireland, be enforced by the courts after a specific period. One such body is the Icelandic Gender Complaints Committee, which has jurisdiction to hear individual complaints submitted by individuals, enterprises, institutions and NGOs, either in their own name or on behalf of their members. The Committee can order that an end should be put to a discriminatory situation, that a remedy should be found and/or that compensation should be paid. Since 2008, the rulings of the Committee have been binding for the parties to each case and may not be referred to a higher authority.33 Previously, if its recommendations were not followed it could refer the case to the courts. Such tribunals might have a tripartite structure, as in the UK, with an adjudicative panel which includes nominees of workers and employers as well as a legally qualified chair. 31 Council of State judgments 4339/1983, 2932/1987, 476/1989 (Plen.) reversing previous case law. 32 Cf. ECHR 27.8.1991 (petitions Nos. 12750/87, 13780/88, 14003/88) Philis v. Greece. 33 Act on Equal Status and Equal Rights of Women and Men No. 10/2008.


A similar adjudicative mechanism, which removes the cost and delay of full-blown court procedures, is the Danish Equality Complaints Board. This body is similar to a court in that it cannot act on its own initiative, either in respect of conducting independent surveys or to start a case. Moreover, the Board is composed of three judges who compose the presidency, and nine members who have a law degree. However, it departs from ordinary court procedures in that cases brought before the Board are decided on the basis of written complaints. If oral evidence (witnesses, party explanations, etc.) is necessary in order to decide the case, it cannot be dealt with by the Complaints Board but only by the ordinary courts. Moreover, Board meetings are not public, and parties to the complaint have no right to be present at the meetings. The procedure before the Complaints Board is cheap and speedy particularly in comparison with court proceedings. It is tax-financed and there is no charge to the complainant. The complaints form can be found on the internet and complainants are supposed to be able to handle their case themselves. If they choose to use professional legal advisers they will have to pay the cost themselves. Similarly, in Romania, claims can be brought before the National Council for Combating Discrimination (the Council) as well as before civil courts. The Council has power to order an administrative sanction (administrative warning or fine). Access to the Council is relatively easy, as no legal representation is required. The Council has power to investigate the facts, organise hearings and decide whether the anti-discrimination provisions have been breached. At the same time, to the extent that these procedures still require that each individual claim be separately processed, it is inevitable that the more they are used, the less efficient they become, unless they are very well resourced. This has been the case both in Ireland and the UK, where the result of large numbers of individual claims being filed has led to significant delays. In Hungary, a claimant has a choice as to whether to seek recourse from the courts or by way of administrative proceedings. Such administrative bodies include the labour inspectorate, consumer protection, health care or education authorities, and the equality body, the Equal Treatment Authority (‘ETA’). As well as having the power to investigate individual or public interest cases upon a claim submitted to it or upon its own initiative, the ETA may also launch court procedures by submitting a public interest claim under the same conditions as NGOs. It also has the right to carry out surveys, to make reports and to assist the Government in decisions on legislative and policy issues. The ETA currently operates under the supervision of the Ministry of Social and Employment Affairs. Nevertheless, it is independent and as such may not be instructed in its actions in investigating or sanctioning discrimination. Its decisions cannot be appealed against within the public administration system, but can only be challenged in the courts. Other models move still further away from a pure complaints-based model by enabling the adjudicative body to initiate its own proceedings. Thus the Bulgarian Commission for Protection against Discrimination can in principle initiate its own proceedings (although in practice, thus far, the primary means of instituting proceedings has been on the basis of a complaint by a victim). Investigation of the case is undertaken by the Commission and the Commission has the power to determine the case, impose sanctions and supervise their implementation. Appeals lie directly to the Supreme Administrative Court. As in the Danish Board, the proceedings are cheap and easy because all expenses incurred in the course of the proceedings are born by the Commission’s budget. Similarly, in Romania, the National Council for Combating Discrimination has powers to initiate proceedings as well as to respond to individual complaints.


A similar approach is that of the Lithuanian Equal Opportunities Ombudsperson. As in its Bulgarian counterpart, the investigation may be initiated by the Ombudsperson on the basis of media reports or verbal or anonymous complaints in addition to a complaint by an individual or legal entity. If the fact of discrimination is proven, the Ombudsperson may a) admonish for committing a violation; b) address an appropriate person or institution with a recommendation to discontinue the actions violating equal rights; c) impose administrative sanctions up to LTL 2 000 (approx. EUR 580); or d) refer the investigation material to the prosecutor if features of a criminal act have been established.34 Where the Ombudsperson chooses to make use of its power to initiate proceedings, rather than react to a complaint, it can markedly increase the use and impact of discrimination law proceedings. In 2002, as many as 34 such investigations were initiated by the Ombudsperson. This increased the total number, including those in response to a complaint, to 72 that year. In 2003, 15 out of 50 investigations were so initiated, and in 2004, 17 out of 57. Since then, however, only a handful of investigations have been initiated by the Ombudsperson and the overall number of gender-related investigations has dropped. A more radical departure from the complaints-led model would be one which emphasises conciliation, rather than an adversarial approach in which the complainant is required to prove that a perpetrator has breached anti-discrimination laws. Several jurisdictions have therefore attempted to establish specialist equality bodies which do not have the power to issue binding decisions, but instead function as a ‘soft-law’ alternative to the court, with the aim of mediating and conciliating between the parties. A particularly sophisticated version of this approach is the Austrian Equal Treatment Commission. A person who feels that she has been discriminated against may of course bring a direct claim to the labour court, but she could also choose to approach the Commission for an opinion. Because the aim is conciliatory, the Commission differs in important respects from a court. Firstly, it is a tripartite body, consisting of four representatives each of management and labour, and three State representatives, one of whom holds the chair. Notably, however, they do not act in a representative capacity, but must be impartial. Secondly, the aim is to help parties understand the opposite point of view and to mediate solutions which could not easily occur in adversarial court proceedings. Thus, the Commission’s initial response is to conduct mediation talks with all parties. Thirdly, the objective is to discover the truth, rather than leaving it to the parties to decide which facts are in dispute. This means that there is no cross examination of parties. Whereas in court the burden of proof is on the complainant to provide the necessary information and the case must be withdrawn if employer shows there were other non-gender reasons for decision, the Commission aims to pursue the truth regardless of the burden of proof. Particularly useful is the power to ask an employer to submit a report containing a detailed comparison between the numbers of men and women working in particular areas, their working conditions, and the relationship between vocational training and job possibilities. The Commission meets in closed session and there are no fixed procedural rules, although the Commission has developed rules in practice. It can decide on an individual issue, or a general issue concerning discriminatory practices, or a group issue such as collective agreements. Its opinions on general issues must be published.

34 Section 169 of the Criminal Code of 28 October 2002 establishes the criminal liability of the natural

person who has committed discriminatory action. The discrimination on the grounds of inter alia sex shall be punished with public work, arrest or imprisonment until 3 years.


A similar body is the Dutch Equal Treatment Commission, which has the power to investigate and adjudicate complaints, and also to initiate an investigation if there is an indication that there is ‘systemic discrimination’. No legal representation is required and the procedure is cost-free. On the other hand, its decisions and recommendations are not binding and it has no power to impose sanctions. The Slovak National Centre for Human Rights has similar powers to prepare non-binding expert opinions and recommendations on compliance with the principle of equal treatment, as has the Slovenian Advocate for Equal Opportunities for Women and Men. In Greece this function is performed for private sector employees by the Labour Inspectorate, which has wide investigating powers, but is not an equality body in the sense of the equality directives, as it is a government service. These include workplace inspections, which can be initiated following complaints or proprio motu, at any time of the day or night, without advance notice; as well as access to undertakings’ records or any other data. It is the only body that has the power to impose sanctions such as fines for infringements of labour legislation.35 Its decisions are subject to appeal to the administrative courts. The Labour Inspectorate is required to cooperate with the Ombudsman. In France, at the end of 2004, a new institution was created: the High Authority against discrimination and for equality (Haute Autorité de Lutte contre les Discriminations et pour l’Egalité, Halde). The Halde is an independent administrative body and it has already demonstrated its capacity to play a very active role in the fight against discrimination. The mandate of the Halde covers all forms of direct and indirect discrimination prohibited by French legislation or in international agreements ratified by France. It plays an important role in supporting individual claims. It may, at its own initiative, investigate cases of discrimination brought to its knowledge without identifying a concrete victim. It assists any and all individuals who turn to it in identifying discriminatory practices and countering them. It holds investigative powers to enquire into cases. It may demand documents and proof which the victim was unable to obtain; ascertain facts on site; and take evidence from witnesses. It provides advice on legal options and helps establish proof of discrimination. It has the power to refer cases to the court system itself on any discriminatory practice brought to its knowledge. The Halde helps identify the procedure best suited to the situation: – A mediation session can be arranged in order to reach an agreement, or when

discrimination has been ascertained, the national prosecutor may be called upon for a decision;

– It may secure compensation, suggest payment of damages to the party discriminated against and trigger proceedings if damages are refused (settlement with compensation);

– It may speak before the judge if the victim decides to go to court; – It may publicly disclose a discriminatory practice. The Halde’s decisions are reached through deliberative proceedings. Each case is investigated to clarify the facts. Decisions present a legal analysis of the situation from the standpoint of anti-discrimination law and set out the Halde’s findings. Its deliberation proceedings include the results of the case investigation. Recommendations, depending on the situation can include: declaring the validity of

35 Including those provided by Act 3488/2006.


the claim; recommending verification measures on-site; proposing a settlement, mediation or referral to court; and providing general observations. In addition, drawing on its experience of the complaints lodged, which can expose inadequacies in legislation or texts which are such that their enforcement can lead to discriminatory situations, the HALDE can issue recommendations so that the legislative and regulatory changes necessary to improve the laws can be adopted. 3. Impact of modified complaints models Modified complaints-led models make an important contribution to addressing the limitations of a pure complaints led approach. This is particularly true where collective complaints can be brought and where forward looking remedies may be triggered by individual or group complaints. Nevertheless, although modified complaints-led models deal with a larger number of complaints than judicial procedures, available statistics show that they are still not widely used. This is true both for bodies which have binding adjudicative powers and those which do not. In Bulgaria, the Commission for Protection against Discrimination, which has judicial powers, finalised a mere 3 cases in 2008, of which 2 were successful. A total of 10 gender discrimination claims were dealt with in 2008. Similar numbers were reported in 2006–7 with the same success rate. Similarly, between 1999 and 2008, gender-related claims before the Lithuanian Equal Opportunity Ombudsperson averaged at just under 48 per year. As the Lithuanian expert points out, this is very low for a country of 3.3 million inhabitants. In Romania, a similar pattern is evident. Although claims before the Council have been increasing, from 11 in 2006 to 22 in 2007 and 32 in 2008, the overall number remains extremely low relative to the size of the population. Notably, of the 26 individual discrimination petitions submitted in 2006,36 more than half (14) were submitted by men. The Icelandic Complaints Committee fares somewhat better, having issued over 160 substantiated opinions since 1995, with 87 of these between 2001 and 2008. The success rate was on average about 31%, although notably in the years since 2005, this rate has dipped markedly. (This is one of the few countries which keep statistics on success rates.) Even at the upper end of the spectrum, numbers of claims remain low, both as a proportion of total referrals, and relative to the size of the population. In Ireland, in 2007, claims of discrimination on the gender ground alone under the Employment Equality Acts 1998–2008, Equal Status Acts 2000 - 2008 and the Pensions Acts 1990–2008 represented 85 out of 852 referrals brought to the Equality Tribunal.37 Of the 113 cases brought under multiple grounds a complaint of gender discrimination was included in 84 such referrals. In addition 33 claims were referred to the Equality Tribunal’s mediation service.38 Although many jurisdictions have had fewer claims in a decade than in Ireland in a year, this still significantly under-represents the scale of inequalities between women and men in Ireland. Similar trends are exhibited in Hungary: – In 2005, the first year of activity, 491 complaints were received by the ETA. It

rendered 144 decisions, finding discrimination in 9.

36 The remainder were submitted by legal entities. 37 Equality Tribunal Annual Report 2007, http://www.equalitytribunal.ie/uploadedfiles//

_AR15July.pdf, accessed 31 August 2009. Equality Tribunal Mediation Review 2007, http://www.equalitytribunal.ie/uploadedfiles/AboutUs/%20Review%202007%20Final%20Proof.pdf, accessed 21 September 2009.


– In 2006, the number of complaints increased marginally to 592, with 212 decisions, discrimination having been found in 27 cases.

– In 2007, the number of claims jumped to 756, with the number of decisions on the merits dropping to 96 cases. Out of the remaining claims, 96 were closed by transfer and 348 with an information letter. Of the 96 cases decided, the ETA found discrimination in 29. Of that 29, only 2 concerned discrimination on the basis of gender.

– In 2008, 1153 claims were submitted to the ETA. The vast majority of them were transferred to another organization or authority or responded to by an information letter to the complainant. Decisions were made in 356 cases, with 22 having been settled by agreement. Out of the 356 decisions, the ETA found a violation of equal treatment in 37 cases. Looking specifically at gender-related claims, 6 decisions found discrimination and 6 did not. The courts’ homepage reflects a similar proportion of gender discrimination related cases- about one tenth of the total (6 out of a total of 59 since 2007).

– As shown above, successful cases before the ETA limited to gender-related39 discrimination numbered 1 in 2005, 3 in 2006, 2 in 2007, and 6 in 2008. Before the courts 4 out of 6 were successful, mostly when hiring or re-employment was rejected.

– The ETA and other administrative organs can only provide limited remedies. The ETA may order the termination of the discriminatory circumstances and prohibit such conduct for the future, but cannot award any compensation or other individual remedy to the aggrieved person. Its main sanction is the imposition of fines from fifty thousand to six million Forints (approx. EUR 200 to 24 000) and order the publication of the decision. Sanctions can be applied by the ETA in a cumulative fashion.

– In addition to receiving a fine, an employer guilty of discrimination can be precluded from obtaining, for two years from the date of the final and binding decision, a so-called ‘orderly workplace’ certificate. Such a certificate is a formal precondition for applying for state (or European) financial resources, and its absence can have grave consequences for the employer.

A salient exception to this trend is the UK, where the number of claims before employment tribunals is significantly higher than in the other countries surveyed: – This is particularly true for equal pay claims, which soared from the already

substantial figure of 4 412 in 2003–4 to 62 706 in 2007–8. – Other sex discrimination claims in the employment field, while not as high as this,

were also substantial, with 17 722 in 2003–4, rising to 28, 153 in 2006–7 and then dipping somewhat to 26 907 in 2007–8.

– Together, sex discrimination and equal pay claims comprised 30 % of all tribunal claims accepted in 2007–8 and 2006–7, compared with 16 % in 2005–6, 13 % in 2004–5 and 11 % in 2003–4.

– Notably, however, a large proportion of these claims are withdrawn (46 % in 2007–8) or settled after conciliation (18 % in 2007–8), and a significant number are struck out without a hearing (28 % in 2007–8).

39 Meaning sex, pregnancy, maternity, family status and sexual identity together.


– A mere 3 % of sex discrimination and 7 % of equal pay claims were successful at hearing. These figures are not out-of-line with the general pattern in previous years.

– Similarly, only a tiny proportion receive monetary compensation: in 2008, compensation was awarded in only 189 cases of sex discrimination, and even then, the average award was only GBP 13 312 and the median GBP 9 109 (approximately EUR 14 700 and EUR 10 000). Reinstatement is not an available award.

– In addition, the tribunal system has had difficulty dealing with very large numbers of claims.

The data themselves do not indicate why the success rate is so low. One possible reason concerns difficulties of proof. There are a number of steps that each individual claimant must take, all of which may be contested by the employer. This particularly true for equal pay cases, where each individual claim requires the identification of a similarly situated comparator who is doing work of equal value; and the employer has the possibility of justifying the pay disparity on grounds unrelated to gender. Similarly, indirect discrimination cases may require the complainant to identify the correct pool of comparison, which may in turn require statistical analyses, all of which may be contested. In addition, the employer may mount a successful justification defence. At the same time, a number of cases (18%) are settled after mediation, and the data do not indicate how many of the withdrawn cases have been settled voluntarily. Equality bodies which do not have binding powers are also not widely utilised. For example, in Estonia, there has only been one application since 2004 to the Chancellor of Justice, who has power to carry out conciliation proceedings between private parties. The Gender Equality and Equal Treatment Commissioner, which has power to issue non-binding opinions, received only 27 applications in 2005–6. This is despite the fact that Estonia has the highest pay gap in the EU. Similarly, in Slovenia, the Office for Equal Opportunities, which can issue non-binding opinions, dealt with only 9 cases in 2008. The Slovenian Human Rights Ombudsman, which has similar powers, dealt with only one case of gender discrimination, which was unsuccessful. This is despite the fact that both these procedures are free of charge. Even in Austria, where there is a network of institutions for the enforcement of equal treatment legislation, relatively few individual cases are noted. In 2004, Senate I of the Equal Treatment Commission for the private sector, which deals with gender equality disputes in labour matters, received 21 cases and made 13 decisions, of which 8 were successful and 3 partly successful. In 2005, 32 cases were received, and 23 decisions issued of which 18 were successful and 2 were partly successful. In France, gender represented only 5 % of the claims registered before the Equality Body or Halde in 2006, 6 % in 2007 and 4 % in 2008. Notably, however, 41 % of these claims were brought by men. Thus, claims by women of gender discrimination clearly remain marginal. The record of the Dutch Equal Treatment Commission, a semi-judicial body, which can issue non-binding Opinions, is somewhat better, in that it received an average of 84 appeals in relation to gender each year between 2004 and 2008, but even this is a relatively small number given the scale of the problem. On average, claims before the ETC are successful about 50 % of the time. It is possible that claims before equality bodies might increase over time, as they become better established and known. There are a few small indications that this


might be the case. In Greece, for example, the Ombudsman was designated as the Equality Body for both the private and the public sector in 2006, and it shows a small increase in individual complaints, up from 11 in 2006 to 24 in 2007. However, the opposite pattern is evident in respect of the Greek Labour Inspectorate, which deals with workers’ complaints in the private sector, and which has a long tradition of monitoring the implementation of labour law as a whole. Here the statistics show only 23 disputes concerning equality in 2006, a mere 0.15 % of the total, which dropped to 11 in 2007 (0.07 % of the total); and 14 in 2008 (0.09 % of the total). The under-utilisation of even modified systems was explained in various ways in the reports. Many mentioned lack of widespread knowledge of these processes by the victims themselves and by union representatives, lawyers, judges and labour inspectors. The Lithuanian expert, for example, contrasted the readiness of workers to complain to State Labour inspectorates in respect of breach of labour legislation, with their reluctance to use discrimination legislation. The latter is attributed in part at least to lack of knowledge of equality legislation or of their applicability to the particular circumstances of the victim. Similarly, in Romania, it was pointed out that although theoretically, the legal system contains the means to address gender-based discrimination, the public perception is that it is complicated, expensive and ineffective. In particular, the mandate of the equality body is not well known. Absence of meaningful sanctions is also referred to as a possible cause, either because the body does not have the power, or because, as in the case of Lithuania, a cautious policy is followed by the equality body in question in the imposition of administrative sanctions on employers. 4. Overall assessment of complaints models and modifications The complaints-led model, when it functions well, is an important avenue of redress for individuals. EU Member States as a rule provide a high standard of substantive anti-discrimination law. Its major strengths are that, unlike proactive measures, it clearly defines who has rights and remedies in cases of sex discrimination. Substantive anti-discrimination law, such as the principle of indirect discrimination, has the potential to address disparate impact as well as unequal treatment. Moreover, as the UK expert points out, by contrast with a top-down enforcement mechanism such as an inspectorate, it does not require official consent or action for the individual to challenge discrimination affecting her or himself. The litigation process might also function as a deterrent to employers or others who may as a result ensure that anti-discrimination laws are respected. However, it is clear that it is not sufficient to rely solely on a complaints-led model. As the study demonstrated, although the possibility of judicial recourse is universally provided, there are remarkably few claims compared to the scale of gender inequality revealed by the figures above. In many Member States, the number of gender discrimination cases before the courts has remained in single figures. Even where there is somewhat greater usage, this significantly under-represents the scale of gender inequality. A constellation of reasons emerges from the reports from the Member States. First, the complaints-led model requires the perpetrator to be identified and proof that the law has been breached. Even with a shift in the burden of proof, it remains very difficult to prove discrimination. Problems of proof are exacerbated in the case of equal pay by the need to find a comparator. This is true too of indirect discrimination, where difficult questions emerge as to appropriate comparisons. A second and particularly significant deterrent is the fear that by initiating legal proceedings, the victim will be exposed or stigmatised as a nuisance.


Thirdly, judicial procedures are slow and, in most Member States, costly. Fourthly, in most Member States, remedies are retrospective and, as the UK expert points out, the amount of compensation available is not sufficient to encourage discriminators to put their houses in order, but rather to wait to be sued and then litigate or settle confidentially. Individuals are required to jeopardise their career prospects, and risk their own resources to pursue a discrimination claim, often without realistic hope of recovery. In addition, in some Member States, there is a low level of awareness among the public in general of discrimination law, and a lack of familiarity among the legal profession itself. The result is that few victims pursue their claims through court, and many breaches go unremedied. It is striking that this occurs even when legal aid is readily available and court costs are not high; or when remedies are real and dissuasive. This suggests that any permutation of the above reasons is sufficient to deter complainants. Paradoxically, too, even when the complaints-led approach is well utilised, its effects are limited, since courts are unable to cope with a large number of individual cases, exacerbating delays. Finally, and most importantly for our purposes, the complaints-led model only addresses discrimination where there is an identifiable perpetrator. In fact, the inequality between men and women has causes which go well beyond individual actions. This means that, in addition to a complaints model, it is important to explore the possibilities of establishing processes which facilitate structural and institutional change. Even when the complaint can be initiated by a union or a class, this model remains limited. There is no direct call for forward-looking or proactive measures, which might bring about more far-reaching change. Moreover, the complaints-led model only addresses discrimination where there is an identifiable perpetrator (which might include the legislator) who can be proved to have breached the law. In fact, the inequality between men and women has causes which go well beyond individual prejudice, and require structural and institutional change. Many jurisdictions have introduced modifications to the pure complaints-led model to mitigate the difficulties faced by the claimant. There are several complementary methods by which modifications have been introduced. One involves lessening the burden on the individual by widening standing rules. A significant number of Member States allow equality bodies, NGOs or trade unions to pursue the case on behalf of the victim, and in some cases, procedures can be initiated even if there is no immediately identifiable victim. Secondly, many jurisdictions have given adjudicative functions to alternative bodies, which are much faster and cheaper than a fully judicial approach. A third approach is to establish alternative dispute procedures which depart from the adversarial model and aim to address complaints of discrimination through compromise and consensus. These processes are much faster, cheaper, and generally less adversarial than a fully judicial approach. In some jurisdictions, they have made an important difference. For example, in Denmark, the Equality Complaints Board and its predecessor have raised the level of compensation for dismissals related to pregnancy and maternity, thereby contributing to improving the protection for pregnant women. Similarly, in Ireland, the expert notes that procedures are relatively straightforward and in general satisfactory. Costs are not at issue; each party bears its own.40 The Equality Tribunal carries out an investigation which in general is very thorough. The possibility of litigation might act as an incentive to employers to adopt their own equality plans. Moreover, the availability of

40 Save where there is a reference on the gender ground to the Circuit Court where costs are at issue,

as many equality claims may be of long duration.


mediation means the parties may resolve the issues early on and the claimant may carry on in the employment concerned. Particularly important, as the Irish expert points out, is the fact that the individual employee or prospective employee does not have to rely on a trade union or the State to enforce their rights. Modified complaints-led models make an important contribution to addressing the limitations of a pure complaints led approach. This is particularly true where collective complaints can be brought and where forward looking remedies may be triggered by individual or group complaints. For example, the non-binding opinions and recommendations of the Dutch Equal Treatment Commission are often followed up voluntarily by defendants and even regularly lead to structural adaptations. Nevertheless, although modified complaints-led models deal with a larger number of complaints than judicial procedures, available statistics show that they are still not widely used. This was explained in various ways in the reports. Many mentioned lack of widespread knowledge of these processes by the victims themselves and by union representatives, lawyers, judges and labour inspectors. Absence of meaningful sanctions is also referred to as a possible cause. In some countries, the full potential of novel procedures has not yet been realised. Thus the Bulgarian expert notes that the modifications of the complaints-led model have not yet been used by the equality body. Nor have newly introduced special rules for judicial proceedings. The effectiveness of these alternative enforcement mechanisms is also highly dependent on resources. For example in Greece, the Labour Inspectorate, to which traditionally private sector workers turn in the first instance, has difficulties in performing its wide tasks, due to insufficient staffing and infrastructure, as its reports constantly deplore. In Ireland, lack of resources and the large volume of claims had led to delays of up to two years from the date of referral to the Equality Tribunal to the commencement of investigation. Equal pay litigation is particularly complicated and time consuming, as has been evidenced in the UK. Alternatively, the equality body might itself be cautious in its use of administrative sanctions, as is the case in Lithuania. Finally, the absence of mandatory sanctions in alternative procedures might be problematic for the victim. Thus in Norway, the expert takes the view that the alternative procedure is paradoxically too efficient so that very few discrimination cases make it to the courts. The result is that both judges and practising lawyers have a lack of knowledge of discrimination law and many people lose out on the compensation they should have received in cases of discrimination. On the other hand, the Polish expert regards it as a weakness of the complaints model that the Polish equality body cannot conduct inquiries in discrimination cases, order to issue binding decisions or award compensation to the victims of discrimination. Moreover, structural change remains elusive, even in the Netherlands, where, as we have seen, the non-binding opinions and recommendations of the Dutch Equal Treatment Commission are often followed up voluntarily by defendants and even regularly lead to structural adaptations. According to the Dutch Equal Treatment Commission, tackling such problems as the gender pay gap demands more than merely enforcing existing non-discrimination law.41 The Dutch report notes that there are no signs that the prevalence of discrimination in the Netherlands is decreasing. For example, complaints about discrimination on the ground of sex, especially with regard to pregnancy and maternity, keep coming in unabated, despite the fact that legislation and case law are very clear on this point.42 As the Austrian expert points out, 41 ETC Annual Report 2008, p. 13. 42 ETC Annual Report 2008, p. 14, available at http://cgb.nl/node/14861, accessed 30 November

2009.


collective means of enforcement are still not as well developed as the severity and dimension of sex-related discrimination would require. Furthermore, the obstacles which hinder the implementation of de facto equality between women and men in society and the economy require more systematic and comprehensive policies. Nevertheless, proactive measures and the complaints-led model should not be regarded as alternative measures. They both have their own peculiar function. The complaints-led model, where successful, makes it possible to satisfy the specific needs of the individual claimant; whereas proactive measures are normally more sensitive to the general needs of the victims of the discrimination. Particularly interesting is the possibility of a creative synthesis between a judicial and a proactive model. This can be created by giving powers to judges or equality bodies to order collective and forward-looking solutions. In Italy, for example, where an Equality Adviser brings a case of collective discrimination to court, the judge can issue an order to the guilty party to carry out a plan to remove the discrimination. Equality Advisers can also propose a conciliation agreement before going to court, asking the person responsible for the discrimination to set a plan to remove it within 120 days. If the plan is considered fit to remove discrimination, on the Equality Adviser’s demand, the parties sign an agreement which becomes a writ of execution through a decree of the judge. Similarly, in Ireland, a successful claimant can obtain significant redress and at the same time there may be an order of the Equality Tribunal or Labour Court that a course of action should be adopted, for example to change a recruitment process. IV PROACTIVE MODELS More recently, attention has shifted to the possibility of alternative and complementary means of addressing gender inequality. These focus on measures to promote or achieve equality between men and women. attention has shifted to the possibility of alternative and complementary means of making equality more effective. These focus on measures to promote or achieve equality between men and women. Rather than being initiated by individual victims against individual perpetrators, such approaches place the responsibility on bodies, such as public authorities or employers, who are in a position to bring about change, whether or not they have actually caused the problem. The EU itself has given particular priority to such measures, both in Article 2 EC, which makes gender equality an EC mission, and Article 3(2) EC, which specifically states that the Union shall promote equality between women and men. The fundamental nature of gender equality is also highlighted in the Lisbon Treaty. Thus, Article 2 of the EU Treaty as amended by the Lisbon Treaty lists equality between men and women among the fundamental values of the EU, while Article 3(3) includes it among he horizontal objectives of the EU. Moreover, Article 8 of the new EC Treaty (‘Treaty on the functioning of the EU’) repeats the provision of Article 3(2) EC. Under Article 29 of the Recast Directive, Member States are required to ‘actively take into account the objective of equality between men and women when formulating and implementing laws, regulations, administrative provisions, policies and activities in the areas referred to in this directive.’


The essence of these measures is that they are forward-looking, requiring bodies43 to take the initiative rather than merely responding to complaints. For example, they could require a body: – to take the initiative in eliminating existing unlawful discrimination; – to scrutinise and change existing practices which perpetuate or cause

discrimination; – to assess all new policies for their impact on gender equality; – to create specific policies to promote gender equality. There is a great variety of measures of this sort found in the practices of Member States. For example: – Some are based in policy measures (soft law) and others in legislation (hard law); – Some are enforceable through judicial mechanisms, and others are not; – Some include detailed requirements to conduct impact assessments of future

policies, monitor change and produce reports, while others are left to the discretion of policy makers;

– Some involve stakeholders in different capacities and other do not. Generally speaking, these measures go beyond a formal equality model, recognising that change would not be achieved if women were simply treated in the same way as men. They aim instead to change practices and policies and this might include measures which treat women differently to men in prescribed situations. An important dimension of proactive change is to address the way in which women’s continued responsibility for childcare affects their ability to participate fully in the public sphere, including paid work, decision-making positions and education. Proactive ‘family-friendly’ measures themselves need to be carefully assessed to ensure that they achieve substantive change and encourage shared parental responsibility. Otherwise, there is a risk that they might reinforce women’s role. 1. The nature of proactive measures Proactive measures can rectify the deficiencies of the complaints-led model in various ways. The first is to provide more effective means of ensuring that existing anti-discrimination laws are fulfilled. Instead of relying on an individual to bring a complaint, responsibility could lie on the employer or a public body to take the initiative in seeking out instances of unlawful discrimination and to rectify them. Moreover, rather than redressing unlawful discrimination only for the benefit of a particular individual, such an approach would seek to find collective solutions, covering all affected individuals. Proactive measures in respect of pay inequality would fall into this category. Secondly, proactive measures might look beyond existing anti-discrimination law and seek to promote equality between women and men regardless of whether there is evidence that a particular individual has been guilty of discrimination. Such an approach is capable of addressing structural discrimination where no specific individuals are responsible. Family-friendly measures and quotas might fall into this category. A third aim is to prevent inequality from arising. This would require screening of new policy or legislative measures to 43 The term ‘bodies’ is used here to include a wide variety of actors, including public authorities,



ensure that they do not create or reinforce inequalities. In the absence of an individual complainant, each of these measures requires the responsible body to take the initiative in diagnosing the inequality in the first place. It is also the responsibility of the body to shape an appropriate remedy and subsequently to monitor the effectiveness of the remedy. 1.1 Making anti-discrimination law more effective: the case of equal pay Pay is one area in which the weaknesses of the complaints-led model are particularly acute. Despite the fact that the right to equal pay for equal work has been part of EU law since the original Treaty of Rome, there remains a significant gender pay gap throughout Europe. As we have seen, Estonia, which has the highest gender pay gap in Europe, has had very few gender equality cases before the courts. Similarly, Cyprus has a significant gender pay gap which, at 22.8 %, is one of the highest in Europe44. This is despite a very strong set of anti-discrimination laws, and a relatively high rate of women’s employment.45 Proactive measures might therefore constitute an important complementary strategy for tackling unequal pay for women. Nevertheless, it is striking how many Member States have taken no proactive measures in relation to pay. These include Bulgaria,46 the Czech Republic, Denmark, Germany, Greece, Hungary, Italy, Latvia, Lithuania, Malta, Poland,47 Portugal,48 Romania, Slovakia, Slovenia, Spain, and the United Kingdom. Among those Member States that have included equal pay among their proactive measures, several different approaches can be discerned. Equal pay could be addressed at governmental level, as part of a general programme of proactive measures, or at work-place level, where the duty might lie on the employer or on the social partners. There are a number of different elements to a proactive policy. The first stage consists of the collection of data, which can reveal the extent of the pay gap and its causes. In Belgium, assessing gender pay gaps and publishing a yearly report on that issue is one of the main tasks of the Institute for Equality of Women and Men. However, it is not sufficient simply to collect statistics. As pointed out by the Latvian report, although the Ministry of Welfare from time to time makes announcements on the de facto statistical situation with regard to unequal pay, this does not lead to any conclusions concerning the reasons for unequal pay or particular measures to tackle this problem. The next stage, therefore, consists of the identification of possible solutions. In Estonia, the Ministry of Social Affairs has the task of publishing reports on the implementation of the principle of equal treatment for men and women. This task includes an assessment of the gender pay gap and the designing of steps to redress pay 44 This gap has, however, been decreasing since 1995. Although there is no statutory obligation to

collect statistics, the government statistic department collects various information, mostly regarding the pay gap and the participation of women in the labour market. The results show that the pay gap between men and women has decreased significantly since 1995 (29 %) and keeps following a downward trend from 25 % in 2005 to 24 % and 22,8 % in 2006 and 2007 respectively. Analysis and suggestions for reducing the gender pay gap in Cyprus, November 2007, Department of Industrial Relations. Cyprus Statistical Services, Labour Statistics 2007, www.mof.gov.cy/mof/cystat/statistics.nsf, last accessed 14 December 2009.

45 In 2007, this stood at 62.4 % (for ages 15–64). 46 But note that a collective remedy was prescribed in an individual complaint - see below. 47 Although in some enterprises ‘good practices’ in this respect are applied. 48 There is a legal obligation to ensure that pay differences between men and women are due to

objective criteria and that job evaluation also relies on objective criteria. That said, there is no specific rule regarding the screening of these criteria, either by the employer or the employees’ representatives or by the inspection services.


discrimination. Relevant studies will be carried out under the 2008–10 Programme to promote gender equality. Similarly, in Iceland, where pay equality is part of declared government policy, a working group has been established to come up with solutions to correct the gender-based pay gap. The current proposal is in the form of a roadmap, leaving it to the discretion of firms to choose a particular mechanism. In Ireland, one of the objectives identified in the National Women’s Strategy is to narrow the gender pay gap. In accordance with that objective, a number of actions have been identified, including: facilitating improved research into gender pay gaps and international best practice; supporting the National Framework Committee on Equal Opportunities; ensuring effective enforcement of the National Minimum Wage Act 2000; and extending equality audits to consider the gender pay gap. Regrettably, however, action has been stalled owing to the absence of funding. Even more targeted is the approach in the Netherlands, where the Government has committed itself to several targets and measures concerning women’s participation in the labour market and narrowing the ‘pay gap’. This is a self-appointed responsibility, based on the political commitment of the Government and the Parliament. Still more specific is the recommendation by the Norwegian Equal Pay Commission that NOK 3 billion (approximately EUR 350 million) should be awarded to increase the pay of women in female-dominated fields of employment who are victims of a lack of equal pay for work of equal value. The Commission saw this as the only way of improving the position of these groups, who are primarily employed by the State and municipalities, given that it is the tough negotiating stance of the State which has led to a significant pay gap. At the other end of the spectrum is the light-touch approach in Liechtenstein, which concentrates on raising awareness. Here the Gender Equality Office, in cooperation with the NGO Infra and representative interest groups of employees, has led an awareness-raising campaign concerning the Gender Equality Act.49 Possibly more effective are those measures which tackle pay inequalities by concentrating on pay structures at enterprise or occupational level. It is here that the primary causes of the gender pay gap are found. For example, in Cyprus, the gender pay gap arises in part from the fact that in many collective agreements, salaries are grouped together into categories A and B, where the former consists of jobs predominantly held by men, and the latter of jobs almost exclusively held by women, resulting in women receiving lower pay. The persistence of unequal pay can be partly attributed to the absence of any obligation to screen job evaluation schemes. Similarly, in Italy, although job classification in collective agreements is ostensibly neutral, it is based on implied factors (such as professional skill and qualification, seniority requirements, and professional experience and capability) that potentially conceal pay discrimination. In addition, the Italian expert points out that although no specific studies are available, pay discrimination could easily be hidden both in additional wages bargained at local or enterprise level and in personal bonuses. This demonstrates the necessity for a duty on employers or the social partners to take positive steps to narrow the pay gap by conducting gender pay audits and correcting any inequalities which these reveal. A very tentative step has been taken in the UK. Here, relevant public authorities, as part of their duty to produce a plan to promote equality of opportunity, must ‘consider the need to have objectives that address the causes of any differences between the pay of men and women that are 49 Created in 1999 and revised in 2006 http://www.llv.li/amtsstellen/llv-scg-gleichstellung-

veranstaltungen/llv-scg-gleichstellung-kampagne_zum_gleichstellungsgesetz-2.htm, accessed 18 November 2009.


related to their sex’.50 There is, however, no requirement to perform or screen job evaluation schemes or otherwise to redress any pay discrimination. A more robust approach is found in Sweden, where Chapter 3 of the 2008 Discrimination Act includes far-reaching requirements for periodical action plans for equal pay by employers with 25 or more employees. (It is worth mentioning, however, that these requirements were weakened by the introduction of the 2008 Act. Now plans are required every three years as compared to every year and the threshold as regards the number of employees has increased from 10 to 25.) Trade unions have a duty to co-operate. Similarly, in Finland, private employers, as part of their positive duty to promote gender equality, must foster equality between women and men as to terms of employment, especially pay. Employers with 30 or more employees are required to formulate an annual equality plan51 in cooperation with the representative of the employees. The plan should contain a report that describes, inter alia, the placement of women and men to different tasks, their job classifications, their pay, and relevant pay differentials. It must also plan necessary steps to promote pay equity and assess the results of previous measures. Hungary adopts a similar approach with respect to public employers.52 The Equality Act requires such employers to adopt an ‘Equal Opportunity Plan’. The purpose of the plan is to eliminate disadvantages of certain groups, such as ethnic minorities, the disabled, single parents, parents with multiple children and employees over fifty. That said, Hungarian law imposes no formal obligation upon trade unions to develop analogous plans. The result is that trade unions in Hungary are relatively indifferent to issues relating to gender equality. Other Member States give collective bargaining a central role in addressing pay inequality. This is particularly true in France, where collective bargaining, at professional branch and enterprise levels, was designated by the Génisson Law of 9 May 2001 as the main tool used to promote gender equality. This law requires gender objectives to be integrated into general negotiations on wages, working time and organisation of work. Equality must be taken into account in every topic of bargaining. The law also creates a specific duty to negotiate on occupational gender equality at branch and enterprise levels. At that level, the employer has a duty to negotiate with trade unions in order to define the objectives concerning equality between men and women in the enterprise and to design the measures to be implemented in order to attain these objectives. In 2006, another law was adopted specifically devoted to the reduction of wage disparities. As well as extending and reinforcing existing provisions, the 2006 Law specifies that the gender gap must disappear before 31 December 2010, referring to collective bargaining for the accomplishment of this aim.53 Collective bargaining is also central in Luxembourg, where any collective agreement must contain provisions consistent with the principle of equal pay for women and men. In addition, the Government has announced its intention to reinforce the obligations that social partners must fulfil in the negotiation of collective agreements. Specific measures aimed at encouraging employers to ensure equal pay will also be instituted during the present legislative period.

50 Sex Discrimination Act 1975 (Public Authorities) (Statutory Duties) Order 2006. 51 Or every third year by local agreement with the employees' union. 52 ‘Public employers’ includes private enterprises 50 % or more of which are owned by the State. 53 See further S. Laulom ‘Gender Pay Gap in Prance’, European Gender Equality Law Review

No. 1/2009 pp. 5-15, http://ec.europa.eu/social/main.jsp?catId=641&langId=en, accessed 30 November 2009.


In other Member States, although there is no legislative compulsion, the social partners have nevertheless taken the equality responsibility onto their own shoulders. In Iceland, trade unions, such as the Commercial Workers’ Union, have equal rights’ policies with an emphasis on equal pay, which include campaigns, advertisements in the media, special seminars for women, and surveys regarding pay equality. In Liechtenstein, collective bargaining has a function in private law rather similar to the law itself, capable of overriding the individual labour contract and applying to third parties. A third of all collective agreements of this sort explicitly contain a clause concerning equal opportunities between men and women, including equal pay. This means that collective agreements referring to equal opportunities between men and women including equal pay (such as in the metal industry, the non-metal industry and the building trades) cover the majority of employees. In the Netherlands, the social partners (trade unions and employers’ organisations) have committed themselves to pursuing the Government’s target of halving the existing pay gap54 by the end of 2011 and to take all possible measures to abolish unequal pay. In addition, the Government has planned to enter into agreements with the Social Partners, and with other relevant parties (e.g. governing bodies of universities, as the pay gap has proved to be considerable at Dutch universities). Some Member States focus instead on providing training for social partners. In Belgium, for example,55 a project (‘EVA’) has been underway for several years to train trade unions’ and employer organisations’ representatives in gender-neutral job classification. This approach has been relatively successful. Although there is no obligation in Belgium for an employer to have a job classification system at all, the latest biennial framework agreements reveal that they are attentive to this issue. When collective agreement n°25 on gender equality in pay was updated in 2008, some pains were taken to make it clear that job classification systems must be made gender neutral. It is also possible to use the individual complaints model to achieve collective change. For example, in Bulgaria, the Commission for Protection against Discrimination ordered an enterprise which practised unequal pay to include equal pay provisions in the collective agreement and to ensure equal pay in future according to the collective agreement or internal regulations. Possibly the most comprehensive package is found in Austria, which recognises that it is not sufficient to focus on pay itself. In addition, measures are necessary to reduce job segregation, to encourage women to move from part-time to full-time work, and to improve the representation of women in higher grades. One of the key features of Austria’s programme of proactive measures is the Austrian National Action Plan on De-facto Equality of Women and Men, which the Government is required to produce in cooperation with the social partners for a period of 5 years. Among the required measures to combat the gender pay gap are the following activities: – The continuation and extension of initiatives aimed at encouraging girls and

women to choose quality jobs and quality vocational training in non-traditional professions;

– Awareness raising of employers for non-traditional career choices of women;

54 The uncorrected pay gap is 14 %, the corrected pay gap (discounting part-time work of women etc.)

is about 4 % in the Netherlands. The target is to halve the pay gap after correction in 2011. 55 At the Federal Government’s initiative, under the Institute’s supervision, and with European Social

Fund’s subsidies.


– Joint initiatives with the social partners for the elimination of hidden forms of discrimination against women in collective agreements;

– The elimination of stereotypes in the assessment of work; – The promotion of transparency of careers and salaries in enterprises; – Initiatives in cooperation with the social partners aimed at supporting the return

from part-time to full-time work and the promotion of women in top positions; – The continuation and extension of mentoring programmes. 1.2 Preventing inequality: impact assessment In addition to eliminating existing unlawful discrimination, proactive measures aim to prevent inequality arising in the first place. A key means to achieve this is to require decision makers to assess new policy or legislative measures to determine their impact on gender and to adjust them if necessary. A significant number of Member States have not, however, instituted impact assessment of this sort: there is no such assessment in Cyprus, France, Greece, Italy, Liechtenstein, Romania, or Slovakia. An attempt was made in France to include impact assessment as part of a law adopted on 27 March 2009 dealing with the parliamentary procedure regarding the adoption of laws.56 The National Assembly sought to include, as part of this procedure, an evaluation of the impact of proposals in terms of equality between men and women before the relevant law was adopted. However, the Senate refused to adopt this amendment and the final text no longer refers to equality between men and women. In other Member States, increasing attention is being paid to impact assessment in order to ensure that legislation and policy measures do not create new inequalities. There are several important features of impact assessment. One is to ensure that responsibility for impact assessment is properly allocated. In a number of Member States, this responsibility lies with Government and the executive. In Belgium, the ‘Gender Mainstreaming Act’ of 12 January 2007 designed a complete system for ex-ante assessment of impact on gender. One civil servant is in charge within each Federal Ministry; there is an interdepartmental monitoring unit; and the Equality Institute provides training and expert assistance. In Austria, initial responsibility lies on all ministries, who are obliged to examine whether and how proposed legislation might impact on the constitutional aim of de facto gender equality. In addition, the directorate on women’s policy, which is situated under the competence of the Minister for Women’s Affairs, is responsible for checking whether and how proposed legislation affects the question of gender equality and delivering comments and amendments to the other ministries. The Constitutional Service Directorate then has the responsibility of assessing whether ministries have indeed examined the question. In Iceland, the Ministry of the Prime Minister has the task of publishing an impact assessment identifying the impact on gender of draft laws and official policy making, while in Ireland, the impact of all new policies is assessed by the Working Group on Equality Proofing. The Netherlands has one of the most developed programs. All Ministries are committed to ‘gender mainstreaming’: this is set up as a standard methodology to identify and assess the (possible) gender aspects of all new policies. In several Member States, however, only a limited commitment is made to impact assessment. In Latvia, for example, the only concrete manifestation of the statutory commitment to mainstreaming57 is a legislative measure58 requiring that draft

56 Loi organique n° 2009-403 relative à l’application des articles 31-1, 39 et 44 de la Constitution. 57 Statutes of Ministry of Welfare.


measures be assessed for their impact on gender equality. In Germany, the rules of the Federal Government require ministers and the Chancellery to carry out gender impact assessments. That said, this obligation is phrased such that these entities should (‘sollen’) rather than must (‘müssen’) carry out such assessments. Additionally, no such obligation extends to public authorities (those entities created under German administrative law), nor private enterprise. Gender mainstreaming is regarded as a matter of good practice in Malta. Responsibility for impact assessment lies with the equality body in several Member States. Thus in Poland, the equality body enjoys general competence to assess existing and new policies in order to determine their impact on gender. Similarly, in Portugal, the equality body can also give advice on the impact of legislative measures with respect to equality issues.59 The position is similar in Malta. In other Member States, responsibility extends to the private sector. In Spain, in order to develop Equality Plans in private companies, a prior assessment of the situation of effective equality in the company must be performed. Once this diagnosis has been conducted, equality goals can be set. Effective monitoring and evaluation methods of the results obtained through application of the Equality Plan must also be devised (Article 46 LOI). In the Czech Republic, impact assessment is a by-product of a more general requirement, namely that when drafting a law it is obligatory to assess the proposed norm from the point of view of international agreements and EC law and to provide information on whether the proposed norm is compatible with EC law. If the norm may affect gender equality questions, its compliance with the relevant gender equality directive must be assessed. There is no obligation to assess the future impact of policies. In other Member States, impact assessment is not obligatory, but could well be chosen as part of the more general duty on mainstreaming. Denmark is an example. A second stage is to ensure that further action is taken once a negative impact has been identified. Not all impact assessments make sure this happens. Thus, in Lithuania, gender mainstreaming is mainly understood as a review of existing legislation with the aim of abolishing any legal provision that is not in conformity with the principle of non-discrimination. However, there is no requirement to review the results of implementation of policies and practices in relation to impact upon gender equality. Public institutions are not obliged to consider the impact upon gender equality, to identify gender equality goals, to create an action plan to achieve these goals or to monitor whether the goals are being achieved and periodically review action plans. By contrast, in Estonia, the Gender Equality and Equal Treatment Commissioner and the Ministry of Social Affairs have the duty of analysing the effect of legislative acts on the position of men and women in society. The Commissioner then has responsibility for making proposals to the Government, government agencies, local governments and their agencies for amendments to legislation.60 A third stage of impact assessment is to review changes once instituted. In Slovenia, impact assessment and review are central to the mainstreaming programme. The Resolution on the National Programme for Equal Opportunities for Women and Men, 2005–2013 (hereinafter the National Programme) is a strategic document which defines objectives and measures as well as key policy makers for the promotion of

58 Instruction of Cabinet of Ministers on drafting of legislative documents binding to institutions of

executive power only. 59 Article 496 No. 1 of Law 35/2004, from 29 July 2004. 60 Article 16 of the Equal Treatment Act.


gender equality in different areas of life of women and men in the period 2005–2013. Before the National Programme was adopted, all existing sector-based gender equality policies were reviewed and an exact analysis of the situation was made. After the expiry of the current National Programme, the same analysis and assessment of existing and new policies will be made to determine their impact on gender. Similarly, in Luxembourg, it is becoming increasingly common to assess National Plans as well as the effect of laws meant to be amended before proceeding. As an example, the latest ‘Plan d’Action National d’égalité des femmes et des hommes’ (National action plan on equality for women and men), which covered the period from 2006–2008, has been assessed and the next plan will take the results into account in order to improve the impact of the National Plan. In some Member States, there is an initial screening mechanism. Thus in the UK, the first step which must be taken in the design or scrutiny of policy is a ‘light-touch’ impact assessment whereby the authority asks itself whether the policy is likely to have a disproportionate impact on gender equality. If the answer is ‘yes’ a full impact assessment must be carried out, the nature and degree of this impact assessment turning on the policy at issue and the size and significance of any likely disproportionate impact on gender equality. In the Netherlands, there is a similar screening mechanism for policies which should be made subject to a more extended impact assessment. There is as yet little direct evidence of the operation of impact assessment in practice. In the Netherlands, several impact assessments have been performed in respect of new laws, regulations and policies. For example, the Ministry for Finance performed an impact assessment on a new tax system several years ago, while the Ministry for Health and Welfare did the same with regard to the new Social Support and Care Act. In Finland, impact assessment is a part of the most recent provision on equality planning. The Finnish provisions are relatively recent, and the efficacy of the duty to assess the impact of previous planning is yet to be seen. What is clear is that the former, generally defined, duty to promote equality by authorities and employers had limited impact. Impact assessment might also be impeded by the shortage of resources. In Poland, where responsibility for impact assessment lies with the equality bodies, their limited staff and budget coupled with the large number of draft laws and existing legal instruments, mean that they can only conduct impact assessment selectively, mainly with regard to draft acts connected with employment, social security or eliminating family violence. 1.3 Promoting equality We have thus far considered proactive measures that aim to root out and redress existing unlawful discrimination, and proactive measures which aim to prevent gender inequality from arising. This section considers the third type of proactive duty - those which take active steps to promote gender equality. The survey concentrated on two such measures: quotas, and family-friendly measures. 1.3.1 Quotas 1.3.1.1 Lawfulness of quotas The use of quotas or set-asides to promote gender equality is inevitably controversial. If a formal approach to equality is taken, any differential treatment on grounds of gender should be prohibited. On such a view, quotas appear to provide preferential


treatment for women and therefore to breach the basic principle of equality. However, on a substantive view of equality, quotas aim to redress disadvantage. In this sense, far from breaching equality, they constitute a means to achieve it. Under EU law, which clearly only applies within its field of competence, Member States may provide ‘specific advantages in order to make it easier for the under-represented sex to pursue a vocational activity or to prevent or compensate for disadvantages in professional careers.’ 61 A corresponding tension is found at member state level, where, in the absence of a clear constitutional or legislative mandate, quotas might be regarded as discriminating against men. For example, in the UK, quotas in favour of women constitute a breach of the non-discrimination legislation, except in tightly defined circumstances (although some further exceptions may be permitted if the Equality Bill currently before Parliament becomes law). In other Member States, the lawfulness of affirmative action or quotas has been the subject of tension between legislatures and courts. In France, statutory provisions prescribing measures to improve on women’s representation on councils and boards were invalidated by the Constitutional Council, on the grounds that restrictive rules based on sex, such as quotas, were not allowed by the Constitution. The revised Constitution, adopted in July 2008, now states that ‘Statutes shall promote equal access by women and men to elective offices and posts as well as to professional and social positions’. Although this gives a constitutional basis to those positive actions which were previously regarded as unconstitutional, no laws have yet been adopted to this effect. In Slovakia, the controversy surrounding the constitutionality of positive measures has been particularly fierce. Section 8 of the Antidiscrimination Act,62 which made provision for temporary affirmative action (compensatory measures), was struck down by the Slovakian Constitutional Court63 on the petition of the Government. In rejecting the institution of compensatory measures as a whole, the Court reversed its previous approach, which had recognised substantive equality. The Court held that the provision was unconstitutional on the grounds that special positive action, including compensatory measures, constituted more favourable treatment (so-called positive discrimination) of persons on the basis of their racial or ethnic origin. In addition, the provision violated the principle of legal certainty by failing to set out the subject and content of the measures in question, or the criteria by which they could be adopted. Finally, the Court insisted that such measures must be expressly declared to be temporary; if not, a situation might occur where they would unreasonably privilege groups of persons, which would lead to prohibited positive discrimination. A further amendment, providing for temporary compensatory measures, was reintroduced into the Antidiscrimination Act with effect from 1 April 2008. Although the initial draft made it lawful to provide temporary compensatory measures on grounds of sex and race, in the final version, gender was wholly omitted.

61 Treaty of Rome, Article 141(4). See C-450/93 Kalanke v Freie Hansestadt Bremen [1995] IRLR

660 (ECJ); C-409/95 Marschall v Land Nordrhein-Westfalen [1998] IRLR 30 (ECJ); C-158/97 Application by Badeck [2000] IRLR 432 (ECJ); C4-7/98 Abrahamsson and Anderson v FogeLqvist [2000] IRLR 732 (ECJ).

62 Section 8(8) Act no. 365/2004 Coll. of Laws on equal treatment and protection from discrimination in some areas changing and amending other laws (Anti-discrimination Act) as amended it read ‘With a view to insuring full equality in practice and compliance with the principle of equal treatment specific balancing measures to prevent disadvantages linked to racial or ethnic origin may be adopted.’

63 PL.ÚS 8/04-202.


Instead, the provisions apply only to membership of a national or ethnic minority and to social and economic disadvantage. Similar trends in judicial reasoning can also be observed in Hungary, where the Constitutional Court may strike down as unconstitutional any measures aimed at ensuring anything more than mere formal equality. Thus, ‘positive’ measures aimed at redressing historical and real disadvantage are not permitted. This is the case, in spite of Article 70A(3) of the Hungarian Constitution, which reads ‘The Republic of Hungary promotes the realization of equality before the law with measures aiming to eliminate inequalities of opportunity.’ As a result, Article 11 of the Equality Act permits and at the same time cautiously restricts the taking of positive measures. The Constitutional Tribunal in Poland has taken the opposite view to its counterparts in Slovakia and Hungary. The Polish Constitution of 199764 contains a general equality and anti-discriminatory clause (Article 32), and declares that both women and men have equal rights (Article 33). The Constitution explicitly allows for preferential treatment only with respect to maternity.65 However, the permissibility of affirmative action has been extended by the Constitutional Tribunal, to encompass social as well as biological reasons. According to the Tribunal, ‘when the biological and social differences between women and men are especially visible (…) the establishment of such privileges becomes even the constitutional duty of the legislator’.66 Other Member States make express provision for affirmative action or quotas. Thus the 1998 amendment to the Federal Constitution in Austria states expressly that measures aimed at de facto equality between men and women are admissible. In Sweden, affirmative action proper or positive action (positiv särbehandling) is dealt with in Chapter 2 on unlawful discrimination.67 According to Chapter 2 Section 2.2 the prohibitions against discrimination do not apply ‘if the treatment of the person concerned is part of an effort to promote equality [between men and women] in working life and is not a matter of applying pay terms or other terms of employment’. A later section on education68 contains a similar permissive rule related to recruitment in other forms of education than basic schooling. 69 Furthermore, the prohibitions on discrimination in labour-market political activities70 and professional access and activities71 contain an opening for efforts to promote equality between the sexes, whereas the rule on membership in certain organisations72 permits benefits for members of a certain sex provided this constitutes a similar effort to promote equality.

64 Dz.U.1997 No 78 item 483 with amendments. 65 In the context of health protection, the Constitution stipulates that public authorities provide for

special health care to pregnant women (Article 68(3)). Article 71(2) provides mothers with maternal leave both before and after birth.

66 Affirmative action means – in the wording of the Constitutional Tribunal – the introduction of preferential treatment in order to diminish factual inequalities between women and men existing in social life. Compare decisions: Kw 5/91, K 27/99, K 35/99 cited after publication: Konstytucja III RP w tezach Trybnału Konstytucyjnego i wybranych sądów (‘Constitution of III Republic of Poland in the thesis of Constitutional Tribunal and other selected courts’) ed. M. Zubik, C.H . Beck, Warszawa 2008 pp. 182–205. All decisions of Constitutional Tribunal are published on the website of the Tribunal:http://www.trybunal.gov.pl/index2.htm, accessed on 18 November 2009.

67 A ‘background’ rule is found in Chapter 2 Section 16 of the Instrument of Government. 68 Chapter 2 Section 6.1. 69 There are also special provisions on positive action in public law and in the area of higher

education. 70 Chapter 2 Section 9. 71 Chapter 2 Section 10. 72 Chapter 2 Section 11.


These rules contain openings for employers/educators, etc., and are to no extent compulsory. 1.3.1.2 Public employment Quotas in respect of public employment are found in Austria, Germany, , and Spain. A sophisticated example of such measures in the public sector is found in Austria, where the Federal Equal Treatment Act73 requires employers to aim to eliminate any existing under-representation of women. Under-representation occurs when women represent fewer than 40 % of all employees within a particular pay grading or type of function. Every six years all ministers must pass an affirmative action plan for their Ministry, setting binding targets to increase women’s representation. These must be adjusted every second year according to the development of the structure of personnel in the Ministry. There are also provisions for preferential treatment in cases of recruitment and career advancement whenever women constitute fewer than 40% of a particular pay grade or function. In such cases, as long as a woman is as well qualified as the best qualified male candidate, she must be hired or promoted until the 40 % target has been reached. Finally, the law provides for preferential treatment for women with respect to professional training. Germany follows a similar template. In selecting employees, public bodies74 are under a legal obligation to choose the person of the under-represented sex if two applicants for the same position are equally qualified.75 The aim of such a practice is to raise the proportion of women (or, of the underrepresented sex) at all levels of public bodies to 50 %. In addition to this, public bodies are under a duty to invite qualified female job applicants and where possible invite equal numbers of men and women to interview. They are also required to specify positions for which there is an underrepresentation of one of the sexes, encouraging members of the opposite sex to apply. Furthermore, public bodies must not indirectly discriminate by requiring applicants to satisfy particular qualifications, such as the length of uninterrupted service or the applicant’s role as breadwinner. There may be obligations to give positive consideration to family and care-related activities if they are of significance for the duties in the position sought. In other Member States, the requirements for quotas are not as comprehensive as in Austria or Germany, focussing on specific types of appointment. In Spain, the focus is mainly on health services, education and in the area of public cultural and artistic activities. Public Health Services are required to ensure the balanced participation of women and men in managerial and professional positions throughout the National Health System. Similarly, in education, the Ministry must promote the balanced participation of women and men in the supervision and management of schools.76 Finally, public authorities in charge of the cultural management system are required to ensure balanced representation of women and men in public cultural and artistic tenders. In Greece, Article 116(2) of the Constitution imposes an obligation on all state authorities to eliminate inequalities and promote substantive gender equality in all fields, in particular by positive measures in favour of women. As a consequence,

73 OJ No. I 65 /2004, Paragraphs 11–11d. 74 This includes all bodies and enterprises established under public law; enterprises that established

under private law are not covered, even if the State owns a majority. In establishing enterprises under private law, the State is obliged ‘to work towards’ an analogous application of the Gender Equality Law.

75 By laws both at the federal level and on that of the states (Länder). 76 Article 24 (2) LOI.


the law requires that women and men respectively should constitute at least one third of appointees by the State as members of ‘service councils’77 of the public service, local authorities and other legal persons governed by public law. The same requirement applies to the boards of legal persons in the public sector,78 provided appointees possess the necessary legal qualifications. In Belgium, the constitution provides that not all members of public executive bodies may be of the same sex. This applies to all levels. In the absence of binding quotas, several Member States have instituted non-binding targets. Thus in Ireland, under the Programme for Government 2007–2012 a commitment has been made to achieving a 40 % minimum representation of women on all state boards and committees under the aegis of any Government Department. As of 31 December 2007, the participation for women on state boards was 34 %. Similarly, the Dutch Government has set several non-binding targets in respect of employment. In the UK, non-binding targets also apply to all public appointments, with a goal of ensuring that 50 % are of women by 2011. 1.3.1.3 Quotas in the legislature Quotas for women in the legislature are more common than might be expected, being found in Belgium, France, Greece, Italy, Portugal, Slovenia and Spain. Particularly noteworthy in this respect is the fact that in Belgium, France, Italy, and Portugal there are quotas in respect of the legislature but not in respect of employment or education. In Belgium for example, electoral legislation (at federal, federate and local levels) provides that parity of sexes must be assured in all lists of candidates. In addition, the first two candidates on every list must be of different sexes. Given that the whole electoral system is proportional, this is likely to be highly effective. In Spain, the principle of balanced participation of women and men, which requires that persons of one gender should not exceed 60 % or fall below 40 %, applies to all candidacies for members of Congress (general elections), municipal elections, elections in the Autonomous Regions and deputies in the European Parliament. In Greece, at least one third of candidates in local and parliamentary elections must be women and one third men. In France, laws of 2000 and 2007 require parties to put forward equal numbers of male and female candidates for election to the National Assembly. For Senate elections, a strict alteration between men and women on the lists is required. Similarly, laws of 2003 and 2007 require mandatory parity with strict alternation of candidate lists in elections to the European Parliament, regional councils, and municipal councils in towns with more than 3,500 inhabitants. If a party does not abide by the rule of parity at local level and to the European Parliament, the list is invalidated. By contrast, there is a financial penalty for non-compliance in respect of elections to the National Assembly. In Italy, there are quotas in respect of European elections. Article 3 of Act no. 90/2004 states that in all lists of different districts which have the same party symbol, neither of the two sexes may be represented in a ratio greater than two thirds. Financial incentives and disincentives are applied to the parties for observing or infringing this rule respectively. This provision implements Article 51 of the

77 Constitution, Article 103(4): ‘Civil servants may not be transferred without an opinion or lowered

in rank or dismissed without a decision of a service council consisting by at least two thirds of permanent civil servants. The decisions of these councils are subject to recourse before the Council of State.’

78 Article 6 of Act 2839/2000.


Constitution, which states that the Republic shall promote equal opportunities for men and women as regards franchise and eligibility. In Slovenia, this applies to both European and national elections. There are 40 % female quotas on candidate lists that need to be respected for elections in the European Parliament and local elections, and 35 % female quotas on candidate lists for general elections. To increase the representation of women in public administration and justice, the principle of gender-balanced representation must be applied in the composition of working bodies, as well as in appointing or nominating government representatives. The principle is therefore applied whenever the representation of one sex is less than 40 %. In several countries, political parties have voluntarily adopted quotas. Thus in Austria, some political parties (such as the Social Democratic Party of Austria and the Austrian Green Party) have self-binding quotas for their representatives in national and regional parliaments. Similarly, in Cyprus, most political parties have voluntarily included quotas to improve participation of women in their decision-making bodies or their lists of candidates. The same can be said of some political parties in Germany, Greece and Poland. In other countries, non-binding targets, rather than strict quotas, have been set. For example, in Ireland, there are no quotas with regard to female participation in the legislature. However, Paragraph 19 of Schedule 10 of the Local Government Act 2001 requires local authorities to seek to promote the objective of an appropriate gender balance in the making of appointments by it to all committees of a local authority or joint committees or bodies of one or more local authorities. 1.3.1.4 Judiciary and other decision-making bodies Only two countries, Austria and Latvia, reported quotas in respect of the judiciary. Paradoxically, in Latvia, where the quota is concerned with the self-governing bodies of the court system, the aim of this measure appears to be to increase the number of male judges in what has been historically a female-dominated profession. More Member States reported quotas in respect of other decision-making bodies (Belgium, Denmark, Estonia, Finland, Greece, Ireland, Lithuania, Slovenia and the United Kingdom). In Belgium, for example, federal and federate legislation requires both sexes to be represented within any consultative bodies (including, for example, the equality body). Usually, any organisation or association that is entitled to send delegates to such a body must propose a woman and a man for each position, unless this is impossible; and within any consultative body, not more than two thirds of the members may be of the same sex. The same quota of two thirds applies to the managing boards of various public institutions (e.g. the Flemish radio and television agency).79 In Denmark, under the Gender Equality Act, public committees, commissions and similar bodies set up by a minister for the purpose of laying down rules or for planning purposes of importance to society should consist of an equal number of women and men. In Finland, under the ‘quota’ provision in the Act on Equality,80 all state committees, councils and other such bodies that prepare decision making, as well as the municipal authorities and the bodies for cooperation between municipalities, are to have a minimum of 40 % of women and men. Notably, however, this does not apply to elected municipal councils. Similarly, the boards of public bodies or companies whose majority owner is a public body must also follow the quota rule. To ensure that the rule is properly implemented, the provision also requires that all bodies 79 In each case this results from the piece of legislation which is specific to the institution being

considered. 80 Section 4 a.


that nominate candidates for such bodies must put forward a female and a male candidate for all positions. 1.3.1.5 Private sector So far as the private sector is concerned, quotas or reservations are unlikely to be mandatory: only Estonia has put in place such measures in the private sector. In Estonia, Article 11(1) of the GEA provides that, in order to promote equal treatment between men and women, an employer shall act in such a way that persons of both sexes are employed to fill vacant positions. Employers must ensure that the number of men and women hired to different positions is as equal as possible; as well as ensuring equal treatment with respect to promotions. In addition, several Member States provide encouragement and incentives to such quotas. For example, in Spain, the balanced participation of men and women in the management and administration of private enterprises is one of the requirements to obtain an Equality Business Distinction.81 Similarly, the achievement of a balanced composition of the boards of directors of companies exceeding a certain business volume is recommended but not mandatory. An eight-year period is set to achieve such a balance. Failure to comply, however, entails no direct legal consequences. Similar non-binding targets are found in the education sector in Austria and Spain. In the Netherlands, although no binding quotas have been set, there are policy targets in some areas of education with the purpose of at tackling job segregation. In Sweden, target goals are set for women professors, but there is no real quota system with regard to appointments proper. 1.3.1.6 Effectiveness Several Member States note that quota provisions have been the most effective form of proactive action insofar as gender is concerned. In Finland, it was noted that while other mainstreaming provisions have included an inordinate emphasis on planning, rather than execution, the positive action that has made the greatest impact in society is the ‘quota’ provision in the Act on Equality. The Act on Equality contained a provision on equal participation in its original form in 1986, but as no exact minimum was set, the provision remained ineffective. After a clear quota was set for state committees in 1995, the number of women in such bodies started to rise. The provision in its amended form has been even more important for municipal decision making. According to a number of studies, the quota provision not merely increased the participation of women in municipal bodies, but it also helped to reduce gender segregation into what is traditionally seen as ‘male’ and ‘female’ functional areas (male technical and female cultural and social bodies in municipalities). Effectiveness does, however, depend on the nature of the remedy, as well as the political culture. As we have seen above, in France, if a party does not abide by the rule of parity at local level and to the European Parliament, the list is invalidated. By contrast, there is a financial penalty for non-compliance in respect of elections to the National Assembly. Financial penalties have, however, proved to be a weak sanction, with parties preferring to pay the fine rather than to comply with the law. The result is that representation of women in the French National Assembly remains weak.

81 Article 50 (4) LOI.


1.3.2 Family-friendly measures One of the key causes of gender inequality is the fact that women remain primarily responsible for childcare and housework. Thus, true equality for men and women will only be achieved if the responsibility for children can be shared more equitably between mothers and fathers. Equally important is the acknowledgement of the value of children to society as a whole. This means that it is not only parents, but the broader community, including employers, the State and others, who should bear some responsibility. Although a detailed study of work-life balance has already been conducted by the gender network82, we believed that it was important to give a brief sketch in this report of the role of family-friendly measures in any set of proactive measures. We were particularly keen to examine the extent to which such measures aimed at increasing the participation of fathers in the care of their children. While growing emphasis is placed on the provision of childcare outside of the home, measures to facilitate participative parenting by fathers as well as mothers remain underdeveloped and under-utilised. 1.3.2.1 General Almost all Member States include family-friendly measures in their constellation of proactive duties. In most cases, these simply take the form of legislative provision of rights and benefits, invariably including the basic requirements laid down by EU law, namely maternity leave, parental leave and time off for urgent family reasons. But several countries now include family-friendly measures in their package of positive measures. These include a wide range of possible measures. At one end of the spectrum are measures, which focus on imparting information. Thus in Latvia, family-friendly measures currently envisaged by the Action Program 2007–2010 are aimed at the information and education of parents on the necessity of sharing childcare responsibilities equally. They also aim to inform parents properly on available social allowances connected with child-rearing. In Spain, by contrast, the focus is on refreshing the knowledge and skills of women or men returning to work in the civil service. Over the first year, preference for enrolment in training courses is given to women or men returning to work following a maternity leave, paternity leave or an extended leave of absence to care for children or relatives.83 Other measures include flexible working options. In the UK, there are legislative provisions which entitle parents of children aged 16 or under (as well as others with parental responsibility) to request changes to their working patterns to assist them in looking after their child. While this is a right to request, rather than a right to have the request complied with, it is backed up with the possibility of indirect sex discrimination claims (in the case of women) and direct sex discrimination claims (by men who might claim that their requests would have been treated more favourably if they were women). Others take a more holistic view, recognising that in order to genuinely change the division of labour in the home and at work, measures must be wide-ranging. The Netherlands is a particularly good example. Here the Government’s equality

82 European Network of Legal Experts in the field of Gender Equality Legal Approaches to Aspects of

the Reconciliation of Work, Family Life in Thirty European Countries, August 2008, European Commission, Directorate-General for Employment, Social Affairs and Equal Opportunities, Unit G/2, available at: http://ec.europa.eu/social/main.jsp?catId=641&langId=en&moreDocuments=yes, accessed 30 November 2009.

83 Article 60 (1) LOI.


programme covers such matters as improving the quality of childcare; raising childcare compensation; facilitating a market for domestic labour; making childcare more flexible; extending the opening hours of local governmental offices to evening hours and Saturdays; and promoting flexible working and ‘tele-working’ among employers. Similarly, Austria has proposed a wide range of measures in its policy programme for the current legislative period. The system of childcare allowances is to be further developed, in particular by the introduction of calculation factors depending on the (former) income of the parent, as well as by introducing more flexibility and by simplifying the rules for the earnings which are permitted in addition to the allowances. In addition, models of better integration of fathers into the family after the birth of a child are to be developed, entailing new labour law and social protection for fathers on leave. Moreover, parents returning to the labour market after child rearing or care for family members are to be better supported. Finally, reconciliation of work and family life is to be generally improved and parents are to be better informed about the different options. In Italy, by contrast, a general commitment is made. Thus, positive action aimed at facilitating the reconciliation of professional and family life is provided by Article 9 of Act no. 53/2000 on the protection of motherhood and fatherhood. Several Member States extend proactive duties to employers, whether private or public. For example, in Estonia, the Gender Equality Act imposes a duty on employers to create working conditions that are suitable for both women and men and support the combination of work and family life, taking into account the needs of employees.84 Similarly, in Finland, private employers have a positive duty to promote gender equality,85 which includes the duty to develop working conditions so that they are adequate for women and men and promoting conciliation of work and family life, especially in arranging work. In Iceland, the Gender Equality Act requires enterprises and institutes with more than 25 employees, on average over the year, to set themselves a gender equality programme or mainstream gender equality perspectives as part of their personnel policy. Reconciliation of family and work is a central element. In Portugal, the Labour Code establishes several measures to promote a more balanced reconciliation of family and working life between women and men (part-time work, several leaves related to care, protection of pregnant women from dismissal, etc.) and, more specifically, to promote the role of the father within the family. The Portuguese expert stresses the importance of these measures for the proactive approach to gender equality in Portugal, especially in the light of the powerful tradition of the mother as primarily responsible for childcare, while at the same time, the great majority of women have a profession and most of them work full-time. There has also been some input from the social partners. For example, in Finland, the non-transferable ‘father's month’ part of parental leave was decided upon by an agreement of 1998–99. In 2005, some sector agreements extended paid leave to include leave for adoption of a child under 7 years, and extended the right to care for a sick child to persons who lived with the child but were not biological parents. In France, the majority of the specific agreements on gender refer to the reconciliation of professional life and private life. Among the measures included are those relating to maternity leave (e.g. extending leave), or paternity leave (e.g. payment of full salary during paternity leave). Others deal with part-time work or services to parents and

84 Article 11(1)(3) of the Gender Equality Act. 85 Section 6 of the Act on Equality.


several provide for financial help for women who need baby sitters to permit them to benefit from vocational training. One way of promoting proactive family-friendly measures is to provide subsidies or other incentives. In Italy, financial assistance is assigned to undertakings enforcing collective agreements that allow for flexible working for natural or adoptive parents through temporary part-time work, tele-work, home work, flexitime and other measures. In Greece, as part of its project, ‘Positive Action in favour of Women in Medium-Small & Big Enterprises’, the General Secretariat for Equality provides financial incentives to enterprises for setting up childcare centres. It has also signed a Memorandum of Cooperation with the Hellenic Network for Corporate Social Responsibility (HNCSR) and a Protocol of Cooperation with the largest federations of employers’ organizations and the HNCSR, for the promotion of gender equality. These includes measures to facilitate family and work reconciliation, for example by creating more flexible working conditions, re-inserting women after career breaks, and encouraging parental leave take-up on the part of fathers. Tax incentives have been a particular feature in Malta. A similar incentive strategy is found in Portugal where an annual prize rewards companies that actively promote gender equality inside their organisation, for instance, by creating childcare facilities or by granting working time conditions that facilitate the reconciliation of family and working life. In Hungary the Ministry of Social and Labour Affairs organises competitions for private employers who provide the most ‘family-friendly workplace’ or ‘women-friendly workplace’. The published evaluation criteria of the competition focus on substantive measures, offering: choice between work-time patterns; other forms of flexibility; study opportunities designed to accommodate the various life stages of the employees; training; the provision of information; part-time work, tele-work or project work offered during and upon return of employees from childcare leave; and taking children into consideration when scheduling vacation. In trying to meet these criteria some employers offer a broad range of social and welfare benefits, including medical screening, medical checks, keeping contacts with and facilitating the return of parents on childcare leave to their job (revealingly referred to as ‘young-moms’ programmes), and even maintaining or hiring nursery places. Maintaining and establishing childcare institutions are highly scored in the competition. However, competitors - even prize winners - fail to show much interest in this respect. It goes without saying that these plans are not enforceable in any way. The only real commitment to such measures is found in the provisions of collective agreements of some employers. Paradoxically however, employers that are paired with strong unions have not often been successful in winning the competitions. 1.3.2.2 Childcare Provision of childcare is a relatively widespread measure taken to promote gender equality. For example, in Ireland the National Women’s Strategy identifies the expansion of quality affordable childcare as a key element of equalising socio-economic opportunities. More concretely, in Austria, as one of the measures aimed at promoting economic growth and employment, the Government proposes to introduce an obligatory last year in Kindergarten, at no charge, for children before they enter school. The Federal Government intends to spend EUR 70 million on this in 2009 and 2010, after which the impact of this project will be evaluated. In addition, its legislative programme for the coming year includes commitments to improve public childcare facilities by initiatives involving the Federal State, the regions and


communities in order to create maximum synergy. Further qualified childcare facilities will be continuously developed and extended, in particular for children under 3, and for full-time childcare facilities with fewer closing days. In Poland, this extends both to state and private enterprise. Thus, measures include the possibility of using the company’s social fund to build and maintain crèches and kindergartens for employees’ children in the company building. However, in several countries, ambitious programmes are not being realised. In Malta, although the Government has moved with vigour, matching rhetoric with some important action, provision of childcare remains well below the Barcelona targets. Nor has the private sector taken any real steps in this respect. In Poland, budgetary constraints are impeding progress, particularly in the light of the recent recession. In Latvia, the Action Programme 2007−2010 envisages the renovation and building of a number of new childcare facilities/kindergartens. However such measures were inserted into the Action Programme only due to the availability of the EU funding for such purposes and they could be stopped or cancelled due to the lack of co-financing from the state budget. Another concern, expressed by the Latvian expert, is that the economic crisis will lead to a considerable decrease in the birth rate and, as a result, new childcare facilities to be built within the next 3 to 5 years will remain relatively empty due to the lack of children in that age group. 1.3.2.3 Fathers Although several of these measures are available to fathers as well as mothers, the focus remains primarily on the mother. Provision for fathers is generally shorter and carries with it compensation at a level which does not provide an adequate incentive to fathers to make use of it. This is particularly true for paternity leave. While provision for maternity leave is generally of a high standard, both in terms of length and remuneration, the opposite is generally true for paternity leave. In several countries, there is no paternity leave. For example, in Cyprus, there is no paternity leave, although there is provision for 18 weeks’ maternity leave. Where there is a right for paternity leave, it is invariably significantly shorter than maternity leave. In Danish law, for example only pregnant women and women who have given birth can take pregnancy leave (during pregnancy) and maternity leave (the first 14 weeks after childbirth). Fathers and only fathers can take paternity leave but at 14 days after childbirth, it is significantly less than maternity leave. In the UK, although there is a short period of paternity leave, the vast bulk of childbirth-related leave applies to mothers rather than to fathers. The same might be observed with respect to Hungary, which provides fathers with an individual right of a mere 5 days in the case of childbirth. Although most fathers avail themselves of this entitlement, other more general parental leave is taken by almost none. A variation of this theme is to allow the father to utilise part of the mother’s maternity leave. In Poland, the father of the child may, with consent of the mother, make use of part of her maternity leave, exceeding 14 weeks, on the same conditions as women. Where there is a right to a prolonged maternity leave, this possibility also extends to fathers. Uptake of paternity leave is significantly improved if it is paid. Lithuania stands out in this regard. Here, a right to paid paternity leave of up to 1 month was introduced in 2006. In 2007, only 3 085 parents took paternity leave, whilst in 2008 this number almost tripled to 9 185. The importance of suitable pay for fathers taking paternity leave is particularly evident in Estonia, where fathers have a right to take ten working days’ leave during the pregnancy leave or maternity leave of the mother or within two months after the child is born. Until 2008, fathers were entitled only to the


minimum wage during this period. Not surprisingly, fathers rarely made use of the so-called ‘father’s leave’. But this situation changed in 2008, when fathers received a holiday allowance amounting to their average salary (with the maximum being three times the average national wage). Regrettably, due to budget cuts in 2009 resulting from the impact of the global economic crisis, provision of this holiday allowance has been suspended. At the same time, receipt of parental benefit is confined to mothers until the child reaches 70 days of age.86 Several measures are available to both fathers and mothers, particularly parental leave, which is a requirement of EU law. Thus in Danish law, both fathers and mothers can take parental leave on the same conditions. The same is true of Germany. In the UK, as we have seen, the right to request flexible working conditions applies to both mothers and fathers. Non-transferable leave is frequently used to encourage uptake by fathers. In Poland, in 2008, additional non-transferable leave for fathers was introduced ( to be applicable from 2010). The right to take such leave expires in the event that the father does not use it during the first year of the child’s life. Similarly, in Luxembourg, workers who have worked for at least twelve months at the time of birth of the child are entitled to parental leave. Parental leave is an individual right and cannot been transferred from one working parent to the other. The monthly overall parental allowance is paid by the State. Yet even when leave or rights to reduced working time are optional and available to both mothers and fathers, and even when leave provisions are non-transferable, the uptake by fathers remains low, a pattern which is consistent across the vast majority of Member States. This is frequently because parental leave is unpaid, or creates insufficient incentives for fathers. In the UK, one of the major shortcomings as regards ‘family-friendly’ working is the fact that parental leave is unpaid. In the Netherlands, although the duration of parental leave has been extended recently from 13 to 26 weeks, it is still unpaid. This is true even when measures are provided in neutral terms: in Italy, for example, Act no. 53/2000 itself addresses its measures to ‘care givers’. Although there are no statistics on this point, everyday life shows an absolute predominance of the mothers in using these provisions. This is clearly related to the level of available remuneration. A similar practice is seen in Hungary, where although various categories of parental leave are not gender specific, they are low paid. The result is that the uptake among fathers is low. Similarly, in Greece, where parental leave in the private sector is unpaid, uptake is very low. In Lithuania, parental leave, available until the child reaches the age of 3, can be taken by either parent; but the vast majority (97 %) of those who make use of this leave are women.87 The same is true in Malta, where although equal emphasis is placed on both parents, the uptake of parental leave by fathers is notoriously low.88 As is the case for paternity leave, fathers’ uptake of parental leave can be significantly improved by increasing the remuneration available to men. For example, in Belgium three specific variations of the more general career break scheme are family oriented: parental leave, leave to attend a terminally ill member of the family, and leave to attend a seriously ill member of the family. When statistics demonstrated that women were the main users by a crushing majority, the Federal Government’s response was to increase the social security benefits which are paid to compensate for the loss of remuneration, up to some 600 EUR per month (nearly double the amount

86 Article 2(2) of the Parental Benefit Act. 87 The situation gets better very slowly. A few years ago the percentage of men taking parental leave

did not top 1 percent. 88 No up-to-date statistics available.


which is paid for a ‘general purpose’ career break). By July 2009 statistics showed that the three specific forms of career break were being used by 39 416 women and 14 953 men – an increase of 4 980 and 4 275 respectively over the last year. In Bulgaria, further incentives for fathers were adopted in January 2009 for the period after the birth of the child (paternity leave). These augment existing opportunities for fathers for raising a small child until the age of 2 years (paid leave) and for a non-transferable 6 months’ parental leave for a child up to 8 years of age (unpaid leave). Both mothers and fathers are entitled to social security benefits. Since 1 January 2009, the mother, insured for general disease and maternity, has the right to benefits at pregnancy and childbirth for a term of 410 days, 45 of which are before birth. When the parents are married or live together, the father has the right to 15 days’ paid paternity leave upon the birth of the child. Furthermore, 6 months after birth, upon the consent of the mother (adoptive mother), the father (adoptive father) is permitted to use the remaining leave of up to 410 days in her stead. The new provisions have yet to be put into practice and thus their efficacy remains unclear. Added to the low level of payment for fathers are cultural expectations, which militate against fathers staying at home to look after children. For example, in the Czech Republic, fathers who remain at home with children are often perceived as strange people. This bias is reflected in the administrative documentation parents must fill out in sending their children to kindergartens and childcare institutions; the language of such forms tends to be particularly adapted for women. It is not surprising that Czech fathers use parental leave in only 2 % of cases. In Estonia, although childcare, flexible working, and parental leave are available to fathers, the latter constitute only about 5.8 % of recipients of parental benefits.89 In Belgium, paternal leave (of 10 days), which does not entail any loss of pay, remains underused. As from the beginning of 2009, it was made more flexible (it can now be taken during the four months following the child’s birth, instead of during the following month), but deeper obstacles are suspected and the equality body has ordered an academic investigation of the issue. In Norway, the need to improve father uptake has led to much effort being put into designing regulations on the right to leave, in addition to the father’s quota. Changing the gender stereotypes as to dividing the care responsibilities between men and women is seen as an important measure in achieving equal opportunities for men and women. The Equal Pay Commission has also proposed to divide parental leave into three parts, one part reserved as non-transferable for each of the parents and one part to be divided between the parents as they see fit. This proposal has not yet been carried forward by politicians. In all these situations, the danger is that family-friendly measures reinforce gender inequality. For example, in Luxembourg, the Ministry for Equal Opportunities conducted a study on the impact of parental leave in 2002. The study showed that parental leave reflects gender-stereotyped roles close to the ones of patriarchal society. The number of mothers taking parental leave is consistently higher than that of fathers. It also showed that employers were more open to allowing a female worker to take parental leave than a male worker. Such reluctance could be countered, as is the case in Luxembourg, by denying employers the right to refuse full-time parental leave of six months. The Latvian system of incentives also illustrates the way in which, paradoxically, patriarchal patterns can be reinforced by measures which appear to be family friendly.

89 Statistics of the Social Insurance Board, July 2009, available at: http:/www.ensib.ee, accessed

31 August 2009.


According to previous legal regulations, the right to childcare allowance was only available to parents who were on de facto childcare leave. This was changed to make the full childcare allowance available to one of the parents irrespective of whether he or she actually works or is on childcare leave. Moreover the amount of childcare allowance is dependent upon the amount of statutory social insurance contributions the parent makes. Since on average men earn more than women, the incentive for fathers to apply for the allowance was increased. The result was that although childcare leave is predominantly taken by mothers, the proportion of fathers receiving childcare allowance increased from 2 % to 86 %. This has been highly problematic, in that, because social protection is provided to the parent who receives the childcare allowance, a considerable proportion of mothers remain without any social protection after childcare leave. With the advent of the recession and the need to cut the social insurance budget, this fact constituted a good pretext to cut the amount of childcare allowance for actually working parents to 50 %. In the light of statistics showing the reluctance of the young father to use paid paternity leave after the birth of a child, the Latvian Action Programme for gender equality 2007–2010 envisages the popularisation of the use of the right to paternal allowance among men. It also aims to elaborate a model of distribution of childcare allowances which facilitates an equal share of child-raising responsibilities. However, due to the economic crisis, it is unlikely that such a model will be implemented in the near future. Indeed, as we have seen, the right to the childcare allowance is itself endangered. 2. Structure of proactive duties 2.1 Responsibility for taking proactive measures? A key element of proactive measures is the shift in responsibility away from the individual claimant to a body which is in a position to take action to address structural inequalities and to ensure that new policies promote equality rather than exacerbate inequalities. The question of who has such responsibility is therefore of central importance. Several different possible responsible bodies were identified: public bodies, trade unions, and private bodies (including private employers), private service providers, and private bodies with public functions.90 Public bodies had responsibility for taking proactive measures in 13 Member States (Austria, Bulgaria, Denmark, Estonia, Finland, Germany, Hungary, Ireland, Italy, Norway, Spain, Sweden, and the United Kingdom). This might include members of the executive, from Ministers, down to regional public authorities as well as public employers. For example, in Bulgaria, the responsibility lies with state authorities, public bodies and local self-government; in Denmark, public authorities are specifically charged with promoting gender equality and incorporating gender equality in all planning and administration within their respective areas of responsibility. In Belgium, under the Gender Mainstreaming Act of 12 January 2007, the Federal Government as a whole and every one of its members individually are responsible for the implementation of the gender mainstreaming policy. The different ministries (‘services publics fédéraux/federale overheidsdiensten’) are the only public bodies directly involved, although independent public institutions such as the social

90 Responsibility for these purposes includes both legal duties and discretionary powers, as well as

political and delegated responsibility. The study included questions on sources of powers. But the range of different ways in which responsibility can be demarcated in different legal and political systems made it impossible to create sensible categories as to whether there was a legal duty or not.


security agencies may be associated with the process. There are no other statutory provisions imposing positive obligations upon any public authority or private persons or associations. However, the action of all public authorities is supposed to be governed by an unwritten principle of lawfulness, under which they should refrain from taking any illegal decisions, and correct any illegal decisions that have been taken in the past. In some cases, the apparatus of responsibility exists, but has little effect in practice. In Hungary, the Ministry of Social and Labour Affairs is the primary governmental institution bearing responsibility for carrying out the goals regarding social equality of genders indicated in the various governmental programmes. In some cases, other organs are involved (e.g., Ministry of Education, Ministry of Economy, etc.); in others, the generality of the wording adopted in laws and regulations results in no responsible body being indicated.91 Apart from the fact that responsibility at this level is only theoretical, a quick look at the period since Hungary’s accession to the EU in 2004 reveals a gradual deconstruction of the short-lived former bodies and institutions that were charged with proactive measures and the promotion of equality.92 The result is that the current department has but illusionary responsibility. In other cases, responsibility is general and therefore somewhat vague. Thus in Latvia, the Ministry of Welfare and the Ombudsman have very broad and indefinite obligations to take proactive measures, without any detailed obligations. In Hungary, public bodies are among the only entities that attract the obligation to take proactive measures. The Equality Act obliges public employers and employers with a state share of above fifty per cent to adopt an ‘Equal Opportunity Plan’. The same might be said of Germany where public bodies are obligated to generate a Women’s Advancement Plan, laying out proactive measures. A general allocation of responsibility runs the risk that everyone assumes that the responsibility will be carried out by someone else. It may therefore be preferable to specify is clearly more effective to specify the duty bearer specifically. For example, in Austria, the public sector is covered by a network of institutions and responsible persons for the enforcement of equal treatment legislation: ombudspersons (up to seven in every Ministry), working groups on equal treatment questions within the ministries and universities, as well as an inter-ministerial working group chaired by the Minister for Women’s Affairs. In Spain, responsibility for the balanced participation of women and men is split between the Government, which is responsible for management boards of the General State Administration, as well as the public bodies reporting to it;93 the General State Administration that is responsible for the appointment of members and representatives of the various Commissions, Screening Agencies; and Committees94 and public authorities in relation to the

91 For example, the National Development Plan for 1997 to 2003 emphasizes the importance of

gender aspects at planning and that the necessary data has to be collected, using indicators, without indicating the responsible body or person. In another example, the Governmental Resolution on the Social Responsibility of Employers assigned the responsibility to make recommendations to advertising and communication companies to the effect that they feature women without prejudice and stereotypes to several different governmental organs.

92 By way of example, the office of the Minister for Equality was abolished directly following formal EU accession, with the function being incorporated into the portfolio of another Ministry. This subsequent arrangement was terminated after 2006 and now gender equality is assigned to a department within the Ministry of Social and Labour Affairs (Roma and disability matters have a ‘chief department’).

93 Act 3/2007, Article 52. 94 Act 3/2007, Articles 53 and 54.


appointment of senior positions. Overall responsibility for promoting the effective equality of women in the various areas belongs to the Public Administration. In Estonia, state and local government agencies are required to promote gender equality systematically and purposefully, and to change the conditions and circumstances that obstruct the achievement of gender equality.95 Thus, upon planning, implementation and assessment of national, regional and institutional strategies, policies and action plans, these agencies have to take into account the different needs and the social status of men and women and consider how the measures applied and to be applied will affect the position of men and women in society.96 The Gender Equality Act (GEA) provides that the Minister of Social Affairs shall make recommendations concerning the performance of the obligations set out above. The recommendations must be published on the website of the Ministry of Social Affairs.97 The Ministry of Social Affairs is also responsible for organising consultation on matters related to the implementation of the principle of equal treatment, giving instructions for the implementation of the GEA, analysing the effect of the legislation on the position of men and women in society, and publishing reports on the implementation of the principle of equal treatment for men and women.98 In Finland, the Equality Unit of the Ministry of Social Affairs and Health is responsible for equality mainstreaming. In addition, public authorities are all required to promote equality between women and men ‘purposefully and according to plan’ in all their activities.99 They are also required to create such administrative and functional measures necessary to ensure the promotion of the equality of women and men in the preparation and making of decisions. The authorities are specifically charged with changing the circumstances that prevent gender equality. In a significant number of Member States, responsibility for proactive measures also falls on trade unions. This is the case for 11 Member States (Cyprus, France, Greece, Iceland, Ireland, Italy, Liechtenstein, Luxembourg, the Netherlands, Norway, Spain, Sweden). In some cases, this is limited to the duty to impart information to their members on the relevant equality legislation (Greece), or to ensure that the collective agreements between employers and employees do not have provisions which create discrimination on ground of sex (Cyprus). In other countries it is more far-reaching. In Iceland, the Gender Equality Act stipulates that trade unions must deliberately work with employers to achieve equality for women and men in the labour market. Several Icelandic trade unions also have voluntary policies to promote gender equality. Sweden has a similar duty on trade unions to co-operate with employers on active measures to bring about equal rights and opportunities in working life regardless of sex, ethnicity, religion and belief. There is also an express duty in respect of equal pay. Such co-operation may result in collective agreements, but this is not mandatory. In Norway, trade unions and employers’ organisations have an equivalent duty to that of public bodies under the Gender Equality Act, namely to make active, targeted and systematic efforts to promote gender equality in their sphere of activity. In other countries, this is more by way of policy formation and co-ordination, as in Ireland, where trade unions play a role in the National Women’s Strategy Co-ordinating Committee which oversees the implementation of the National Women’s Strategy,

95 Gender Equality Act, Article 9(1). 96 Gender Equality Act, Article 9(2). 97 Gender Equality Act, Article 9(3). 98 Gender Equality Act, Article 22. 99 Act on Equality, Section 4.


which is aimed at providing a structured approach to gender equality promotion. In Italy, a positive incentive is applied to trade unions: national and regional trade unions can promote positive action and ask for financial support as provided by the law. So far as reimbursement of costs for positive action plans is concerned, priority is given to those adopted on the basis of collective agreements bargained between employers and trade unions. In the Netherlands, social partners are only responsible insofar as they have committed themselves in agreements with Government. They have thus far committed themselves to pursue the Government’s target of halving the existing pay gap by 2011 and to take all possible measures to abolish unequal pay ultimately. Interestingly, in Germany, any suggestion of the imposition of proactive duties upon trade unions has been rejected on the grounds that to do so would violate the freedom of social partners (Koalitionsfreiheit). Duties on trade unions are usually matched by duties on private employers, as is the case in Cyprus, France, Greece, Iceland, Ireland, Italy, Liechtenstein, Luxembourg, the Netherlands, Norway, Spain, and Sweden. In a handful of countries, private employers might have duties even in the absence of duties on trade unions. These include Bulgaria and Estonia. Although in Germany, as with other Member States, private employers are not under any legal duty to pursue proactive measures, a trend of voluntary adoption has emerged. In 2001, during the ‘red-green coalition’, employers’ organisations entered into what is known as a ‘Voluntary Commitment’ (Freiwillige Selbstverpflichtung). This Voluntary Commitment contains duties to improve gender equality, and was a response to the Government’s threat to enact a law imposing obligations upon employers to take such measures. Monitoring of this commitment is undertaken by the Federal Ministry for Family, Senior Citizens, Women, and Youth (Bundesministerium für Familie, Senioren, Frauen und Jugend). The success of these plans in achieving gender equality have been controversial. The positive picture drawn by past monitoring reports, has been fiercely contradicted by women's organisations . Private bodies in their capacity as service providers are responsible in Bulgaria, Iceland, the Netherlands, Norway, Spain and Sweden. There were also several other bodies which had identified responsibilities, primarily in the area of education (Estonia, Finland, Norway and Sweden) but also in other respects (Romania, Portugal). 2.2 Consultation In many circumstances, proactive measures can be bureaucratic and ‘top-down’. It is therefore of central importance to involve stakeholders, potential victims, trade unions, service users, relevant NGOs, and others in the process. Consultation raises some complex challenges. One concerns the function of consultation. It might be limited to the imparting of information by policy makers or employers to stakeholders, or it might extend further into an exchange of information and views. Alternatively, it might consist of fully-fledged co-decision-making, possibly in the context of collective bargaining. Also of importance is the subject matter and level of consultation: consultation can take place at several levels, ranging from consultation over legislation to enterprise-level consultation or bargaining. Particularly challenging is the method chosen to identify consultees: in the absence of a specified victim, questions arise as to the representativeness of consultees, their expertise, and their capacity to engage in the process. Ideally, there should be involvement at all stages. Stakeholders should take part in the original diagnosis of the causes of inequality and then play a role in the subsequent development of an equality plan or bundle of


proactive measures. Furthermore, once measures are in place, stakeholders should participate in the monitoring and review of such measures as well as their ultimate enforcement. Consultation of some sort is found in 21 Member States (Austria, Cyprus, Estonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Spain, Sweden, and the United Kingdom). In saying so, however, there is a wide variety in type and function of involvement. 2.2.1 The subject matter The subject matter of consultation can vary significantly. Particularly evident is the growing trend towards canvassing the views of stakeholders on draft legislation or policy measures. For example, in Belgium, the Council for Equal Opportunities between Men and Women was created in 1993 primarily to satisfy requests by ministers for opinions on draft legislation. More recently, this has expanded to include the drafting process itself: the Federal Minister of Equal Opportunities has recently instructed the Institute for Equality of Women and Men to draft the ancillary Royal Decrees for the implementation of the Gender Mainstreaming Act. The legislative review model is also found in Poland, where the right to be consulted is provided in several different legislative acts.100 Regulations applying to the Council of Ministers require all interested parties to be given access to each draft act or regulation in order to present their opinion on its expected effects on gender equality. In practice, draft acts and regulations and all policy programmes are presented for consultation to the members of the so-called ‘Triangle Commission’, which regularly engages in dialogue on important social and economic issues.101 Similarly, in Estonia, the Minister of Social Affairs must consult on matters related to the implementation of the principle of equal treatment, particularly when fulfilling its duty to make recommendations on the effect of legislation on gender equality.102 This has recently been extended further: Article 9(21) of the Gender Equality Act establishes a statutory duty on state and local government to consult with interest groups, if necessary, with respect to the introduction or implementation of new strategies, policies or action plans. Such robust regimes might be contrasted with that of Hungary: although a consultative body, the Council of Social Equality of Women and Men, is entitled to give opinions on draft bills, the expert doubts both the frequency and efficacy of such assessments. Sometimes the subject of consultation relates to all matters concerning the policy initiatives of a given Ministry. In Latvia, the Ministry of Welfare is obligated to consult in adopting any decisions within its competence. Consultation can also occur as part of the process of formulating specific equality plans. Thus, in the United Kingdom, a public authority must, in drawing up its gender equality scheme, ‘consult

100 By way of example, such a procedure was provided in the decree of the President of Council of

Ministers of 13 January 2005 on procedure and terms of consultation of National Programme of Development, operational programmes and strategies of the use of the Cohesion Fund, Act of 6 July 2001, Dz.U. 2005 No. 10 item 74.

101 The Act of 6 July 2001, Dz.U. 2001 No. 100, item 1080 with amendments. In addition to being composed of civil servants, the Commission includes representative organisations of employers and employees. Such ‘representatives’ are typically nationwide with more than 300,000 members or employees.

102 Gender Equality Act Article 22.


its employees, service users and others (including trade unions) who appear to it to have an interest in the way it carries out its functions.’103 Proactive measures at enterprise level also constitute an important arena for consultation, with employment conditions forming a considerable focus of consultation. In France, any enterprise employing 50 workers or more must present a report to the Works Council outlining the relative position of men and women in their operation. The objective is to provide and obtain information, establish a shared diagnostic of the employment situation, and to reflect on measures that might be taken to improve equality. Similarly, in Norway, Working Environment Committees as well as workers’ representatives (elected by the employees) regularly participate in a consultative process in order to determine what equality measures may be useful to introduce and to evaluate their impact. For bigger issues, employers may invite all employees for a mass information/consultation meeting. Where consultation is not mandatory, it is possible for trade unions to take the initiative in this respect. Thus, in Belgium, one trade union has repeatedly (and, as recently as July 2009) taken the opportunity provided by compulsory negotiations with trade unions in the federal civil service to block proposals that might negatively impact rights to maternity or parental leave.104 In Germany, consultation takes place only in ‘public bodies’ – those entities created under administrative law. This category includes public administration on the federal, state and local level (but not elected bodies), universities, state-run schools, state-run hospitals, and health administration (but not private hospitals or private health insurance). The substance of such consultation concerns the development of Women’s Advancement Plans – documents setting out the way in which the number of women on all levels of the organisation is to be increased. Interestingly, proactive consultation can also occur in the context of the complaints-led model. Italy exhibits a complaints-led model that is slightly different from the traditional archetype. As mentioned above, the legal system permits the making of a complaint with respect to collective discrimination in a workplace. In such a case, the judge must consult both the relevant workers’ representatives and Equality Adviser before issuing an order to adopt a plan to remove such collective discrimination. Settlement of such cases can only occur in circumstances where such consultation results in the adoption of a plan to remove the discrimination; such plans must be agreed to by both the employer and Equality Advisor. 2.2.2 The parties A review of the practice of Member States yields numerous and varied examples of entities involved in consultation. One approach is to create a consultative framework, which can respond to particular issues. For example, in Iceland, the Minister of Social Affairs and Social Security may engage a gender equality counsellor to work temporarily on gender equality issues in a specific field or area.105 More durable structures are found in other Member States. Thus, in Ireland, the main forum for consultation is the National Women’s Strategy Co-ordinating Committee, which oversees the implementation of the National Women’s Strategy. The Committee comprises a broad membership including members of each government department and state agency, social partners, representatives from the

103 Sex Discrimination Act 1975 (Public Authorities) (Statutory Duties) Order 2006. 104 That said, this practice is entirely due to the vigilance of one trade union, rather than the existence

of any obligation. 105 Gender Equality Act (n x) Article 14.


National Women’s Council of Ireland and its constituent groups, and one member from the Equality Authority. The respective interested sectors are identified by the Department of Justice itself who then nominates members to the Co-ordinating Committee. As social partners the unions play a role in the Co-ordinating Committee. Additionally, the Inter-Departmental Committee, which drafted the National Women’s Strategy, continues to meet regularly to enable exchange of information between Departments. Romania too has established a framework for regular consultation in the form of quarterly meetings held by the National Commission in the field of Equal Opportunities between Women and Men (‘CONES’). 106 Within the CONES quarterly meetings, reports submitted by the county commission are to be analysed with the aim of deciding on the implementation of concrete measures for the elimination of identified gaps in the implementation of equality for women and men at the local level. The Commission is formed from representatives of the ministries and other specialized bodies of the central public administration subordinated to the Government, trade union and employers’ organization representatives at the national level, as well as of representatives of NGOs active in the field of equal opportunities for women and men. Similarly, in Luxembourg, the ‘Comité du Travail Féminin’ (Committee on Female Employment) is a consultative body, seized by the Ministry of Equal Opportunities. It is composed of representatives from social partners,107 the National Council of Women, and the Government. In Estonia, the Gender Equality Act envisages the establishment of the Gender Equality Council - an advisory body within the Ministry of Social Affairs. Its consultative relationship with the Government will be twofold. It will both advise in matters relating to the promotion of gender equality and present its opinion concerning compliance of the national programmes of the various ministries with Article 9 of the Gender Equality Act.108 This consultative process may require discussion with relevant NGOs and interest groups who have a legitimate interest in preventing gender discrimination.109 The amendment’s explanatory memorandum identifies local women’s organisations and their roundtables, men’s organisations, educational institutions, and other organisations dealing with human rights as suitable consultees.110 In practice, the number of NGOs dealing with gender equality is small. Most are well known to governmental institutions, and their input on draft legislation is regularly sought. However, as foreshadowed, the Council is yet to be established in practice. Other Member States designate potential parties more specifically. We have seen that in the United Kingdom, a public authority must, in drawing up its gender equality scheme, ‘consult its employees, service users and others (including trade unions) who appear to it to have an interest in the way it carries out its functions.’111 In Poland, the Law on public profit organisations guarantees NGOs the right to take part in social consultations.112 Similar rights are also provided in many regulations governing specific professions, associations, scientific institutions, etc. In other Member States,

106 Act on Equal Opportunities 2002 Article 32. 107 The term ‘social partners’ refers to trade unions and employers that are engaged in social dialogue. 108 Gender Equality Act (n x) Article 224(1).. 109 Gender Equality Act Article 9(21). 110 An Explanatory Memorandum to A Draft Act to Amend Gender Equality Act, Labour Contracts

Act and Equal Treatment Act, No. 317 SE http://web.riigikogu.ee, accessed 31 August 2009. 111 Sex Discrimination Act 1975 (Public Authorities) (Statutory Duties) Order 2006. 112 Act of 24 April 2003, Dz. U. 2003 No. 96 item 873.


consultation occurs with a number of entities, including social partners. Thus, in Latvia, the development of gender equality policy involves consultation by the Ministry of Welfare with the Confederation of Employers, the Confederation of Trade Unions, the Ombudsman, representatives from various ministries, and the most prominent relevant NGOs. In Italy, consultation of trade unions, the Equal Opportunities National Committee (‘EONC’), and the National Equality Adviser accompany the adoption of three-year positive action plans in the public sector. Additionally, the EONC is to be consulted on the financing of the Minister of Labour’s positive action plans. Several Member States rely on the equality body as a consultative forum. In Cyprus, the National Machinery of Women’s Rights (‘NMWR’), established under the Minister of Justice and Public Order, plays a central coordinating role in the promotion of gender equality in Cyprus and ensures the introduction of gender mainstreaming in all national policies and programmes. It deals with all matters concerning women, focusing on the elimination of legal discrimination and the promotion of equality. More specifically, it advises the Council of Ministers on policies, programmes, and laws promoting women’s rights. It monitors, coordinates and evaluates the implementation and effectiveness of such programmes and laws, and also disseminates information, organizes seminars and training programmes on relevant issues, and supports and subsidizes women’s organizations. In discharging its advisory role to Government, the NMWR consults widely with its members. Membership includes a number of NGOs dealing with equality issues and women’s rights, and a range of government departments. In Greece, the National Commission for Gender Equality is required to liaise with civil society in the elaboration of gender equality policies in all fields, suggest measures for implementation, and evaluate results.113 The Commission is chaired by the Minister of the Interior, or the General Secretariat for Equality. It consists of twenty members: representatives of six ministries, local authorities, the Economic and Social Council of Greece (‘ESC’), workers’ and employers’ federations, three country-wide NGOs and two experts in gender equality. However, thus far, this body has only met twice and its impact is not visible. The General Secretariat for Equality also consults with women’s NGOs on an informal basis and so does the Ombudsman. The Ombudsman, the ESC and the National Commission for Human Rights (an independent consultative body)114 are often invited to hearings by parliamentary committees elaborating bills on matters relating to their remit. Similarly, in Iceland, the Gender Equality Council, appointed after parliamentary elections, has the duty to advise the Minister of Social Affairs and Social Security and the Director of the Centre for Gender Equality on policy touching on gender equality issues. In Portugal, the Equality Agency in the area of employment (‘CITE’), responsible for a number of initiatives,115 acts in consultation with a number of bodies. In this respect,

113 Established by Article 8 of Act 3491/2006. 114 Established by Act 2667/1998 (OJ 281/18.12.1998) in accordance with the UN Paris Principles

(www.nhri.net, accessed 29 August 2009). NCHR website: www.nchr.gr, accessed 30 August 2009. 115 CITE enjoys broad competence to bring cases of general discrimination. This may range from

discriminatory practices, or discriminatory clauses in labour contracts or collective agreements. CITE assesses the content of collective agreements for any discriminatory clauses to be eradicated by court order. LC, Article 479. CITE does not have the power to bring actions before the Courts as a representative of the victims, but it is an important counsellor in this area and can direct the complaints to the ACT. Also, this Commission must give advice prior to management measures that may have discriminatory implications, such as the dismissal or the non-renewal of a fixed-time labour contract of a pregnant worker, and the refusal to access to part-time work for care reasons.


CITE depends on the Ministry of Labour, representatives of the major national trade unions, and major employers’ associations. These parties deliberate together in assembly. In this sense, this Commission is not directly under the thumb of the Government but is a truly ‘tripartite’ organisation. CITE enjoys a broad discretion to seek the advice of other private or public entities, and experts. In Sweden, the Equality Ombudsman has a duty to inform, educate, consult and cooperate with public bodies, private enterprises, and organisations within its area of activities. Its remit also includes a duty to follow international developments, make contact with international organisations, approach the Government concerning actions against discrimination, and take other ‘appropriate’ action.116 In several Member States, the major consultees are trade unions and workers’ representatives. Thus, in Finland, equality planning by employers must be made in cooperation with employee representatives. In larger workplaces, the employee representative is typically a trade union, which plays a central role in equality planning. Consultees are not limited to worker representatives, but can extend to employers as well. In France, employers with more than 50 workers are obligated to present an annual report on the comparative situation between men and women to the works councils. In Germany, the limited consultation that takes place in public bodies is coordinated by a Gender Equality Ombudsperson. She is elected by the female employees of the relevant public body, and gathers information and provides advice on the development of the body’s Women’s Advancement Plan. Not all consultative fora have been functional or effective. In Latvia, legislation was promulgated creating the Gender Equity Council, an organisation with the aim of improving gender equality policy. The Council was composed of various actors: representatives of key ministries, municipalities, rural folk, social partners, scholars, and leading NGOs. Although the structure showed much promise, the reality was disappointing. Because the results of the consultation were non-binding, the Council’s conclusions were ignored by politicians and policy makers. In light of this ineffectiveness, in 2008 the Cabinet of Ministers adopted a decision to liquidate the system entirely. As such, although these regulations remain on the statute books they no longer enjoy efficacy.117 Similar trends might be observed in Hungary with the establishment of a governmental body aimed at consultation, known as the Council of Social Equality of Women and Men. The Council consists of government executives, experts nominated by the relevant minister, and NGO delegates. The body is entitled to give opinions – including on draft bills, participate in the preparation of support programmes, and make proposals. The body is aimed at encouraging consultation within itself, and in pursuing this objective it seeks to bring together relevant interest groups. The expert points out that the efficacy of this arrangement is questionable. Not only are consultation and its results non-binding, but the Council’s success is bedevilled by historical obstacles. The Council’s pattern of practice reflects the legacy of the corporatist pre-transition political regime, in which social groups and organisations were appointed to senior positions in the Party whilst being accorded no real influence in the decision-making process. In Poland, as we have seen, the law on public profit organisations guarantees NGOs the right to take part in social consultations. Research conducted by the Ministry of Economy proved, however, that in 20 % of governmental draft laws such consultation has not been conducted and almost half of public authorities had not

116 Section 3. 117 Regulations No. 79, OG No.31, 26 February 2002.


invited NGOs for consultation in the first place.118 The Greek Commission has only met twice,119 and there is no general agreement or consensus on its impact. 2.2.3 Selection The selection of consultees is also a very important aspect of the process of consultation. Attention needs to be paid to the representativeness of consultees, their expertise, and their capacity to engage in the process. In particular, the selection process for consultees should aim to bring into the process those individuals or groups best poised to provide critical and constructive feedback in the development of proactive measures. There is a risk that failure to ensure the selection of suitable consultees might jeopardize the entire process. As with other areas, the method for selecting consultees varies widely. In Finland, for example, the manner of choosing those involved in equality planning in educational institutions is given in regulations concerning these bodies; representatives are prescribed under labour law. The same might be said of Italy, in which the relevant bodies (i.e., the EONC, the National Equity Advisor, and trade unions) are permanently charged with the task of reflecting on specific issues. Other variations exist, however, in which controlling effect is given not to regulations but to a government ministry. In Ireland, the members of the National Women’s Strategy Co-ordinating Committee are identified by the Department of Justice itself who then nominates members. In some Member States, there are procedures for the formation of broader consultative committees or commissions. Iceland exhibits selection according to a mixed process. Article 8 of the GEA provides that the Gender Equity Council is comprised of a chairman, appointed by the Minister of Social Affairs, and additional members nominated by various labour, feminist, and women’s rights associations. In other jurisdictions, the selection of consultees is dictated by custom. In Latvia, no regulations lay down any procedure for selection. Instead, the practice is merely such that particular partners are singled out for consultation. Currently, the Ministry of Welfare organizes informal meetings with NGOs on a quarterly basis. Similar trends are exhibited in Luxembourg, where generally speaking the Government invites the most representative stakeholders on any given national plan to provide input. Relevant non-profit associations, trade unions, and employers’ associations typically designate their representatives depending on the subject. This seems to be equally the case in Poland120 Romania,121 and the United Kingdom.122

118 See the analysis of the Ministry of Economics, Ocena skutków regulacji. Zasięganie opinii osób

zainteresowanych stron w procesie oceny skutków regulacji (‘The assessment of regulations’ results and the consultation with concerned parties’) www.mg.gov.pl, accessed 28 August 2009.

119 See www.isotita.gr, accessed 29 August 2009. 120 In Poland, there is no clear criterion with respect to the choice of consultees. In most cases the

scope of consultees is limited and none of the concerned organizations enjoy a legal guarantee to have the opportunity to participate.

121 No specific criteria are available regarding the way in which members of the CONES are selected. Information is available only with respect to the NGO representatives, who are designated by the relevant NGO itself.

122 This applies to the selection of consultees by listed public authorities. Neither the Regulations nor the SDA regulate how consultees are chosen, although on the ordinary principles of public law and in order to comply with its general duty an authority would have to act reasonably in selecting consultees. The Code of Practice suggests (Paragraph 3.29) that: ‘The extent of consultation should be appropriate to the size, remit and resources of the authority and there is no prescribed means of carrying it out. Public authorities are free to adapt their existing processes of public consultation. It is important to remember, however, that the duty is to consult on gender equality. Women and men


2.2.4 Is consultation binding? Whether or not consultation is binding may be answered by reference to two different questions. The first concerns whether or not the act of consultation is itself obligatory. The second concerns the way in which the information or advice provided in the course of consultation must be treated. This question may be further broken down into two separate categories: those consultations that require the recipient to pay regard to the information and advice provided, and those where the recipient is bound to follow the advice provided. In the former category, it is often the case that if the decision maker chooses to disregard the information s/he must provide reasons for doing so. Finally, there may be a wholesale delegation to the consultee of the function of designing a proactive measure. In most cases, consultation is aimed at giving and gaining information, and gathering ideas, but does not entail any binding effect on decision making. Despite this, opinions of consultees contain the potential to have considerable force. No more is this seen than in the case of Greece, where non-binding opinions of the various bodies involved in consultation, such as the ESC, the NCHR, the Ombudsman, and the General Secretariat for Equality, are, in practice, followed. This pattern is reflected in the example of France, where there is no obligation for the employer to take into account the opinion of the Works Councils, but nevertheless the practice is to do so. In the Belgian regime, the making of such a request is not obligatory; ministers are under no compulsion to seek the opinions of the Council. Despite this, its opinions are regularly sought. In the case of Italy, consultation is a procedural step with no binding effects in the complaints-led model. That said, the role of the EONC is crucial, as it is the body that formulates the annual programme fixing the targets for proactive action. The body is also charged and empowered to monitor implementation. In this way, although it might be said that in its consultative role, strictly speaking, none of its conclusions are binding, it has far more plenary powers that entail an ability to reach beyond the mere provision of information. However, the distinction is not merely that consultative opinions are either binding or not binding. In some Member States, the decision maker, although not required to follow an opinion, is nevertheless required to give reasons for such a course. Thus, in Estonia, although the opinions of the relevant interest groups are not binding, the duty to consult implies that the decision maker must give reasons if their opinions are not taken into account. Equally, government regulations require that draft legislation sent for comment be accompanied by reasons for disregarding relevant consultative opinions.123 A different approach is to require the decision maker to take the views of consultees into account, whilst not being bound by them. The consultation obligation placed upon listed public authorities by the United Kingdom is of this nature. No aspects of Ireland’s consultative process with the National Women’s Strategy are binding. The same might be said of the consultative processes of Germany,124 Iceland,125 Lithuania,126 Latvia,127 Norway,128 and Poland.129 To a certain extent this is

(and, where appropriate, girls and boys) should both be consulted, but public authorities will have to ensure that the consultation process gives adequate attention to issues of gender equality, and any questions are structured in such a way as to bring out any potential differences in views between women and men, or between groups of women and men.’

123 Regulation of the Government of 28 September 1999, No. 279 on rules on legislative drafting, Article 38.

124 Any information gathered or input provided by the relevant Public Body’s Ombudsperson need not be taken into account and has no binding effect, respectively.

125 No binding effect is attributed to consultation involving the Gender Equality Council.


true of Slovenia and Hungary as well. There is no specific obligation upon the responsible body to consult others in relation to proactive measures. There is only a general obligation of the Government, ministries, and other state bodies to co-operate between themselves and with social partners and NGOs that are active in the field of equal treatment regarding the elaboration of solutions and proposals to achieve the purpose of the AIPET. That is, to ensure equal treatment of all persons in the assertion of their rights and the duties owed to them, and in the exercise of their human rights and fundamental freedoms. In other cases, it seems possible, at least in theory, for consultations to yield binding proposals, but that none are ever adopted. In Malta, periodic discussions take place in fora such as the Maltese Council for Economic and Social Development (‘MCESD’). However, such discussions consistently fail to render binding proposals for pro-active measures, limiting their memoranda merely to the provision of information and discussion. 2.2.5 Assessment As the above illustrates, the practice, as in other aspects of proactive measures, is extremely diverse. The procedures adopted by the various Member States differ with respect to the subject matter of consultation, whether it be in relation to the generation of policy plans, the vetting of legislation, or the modification of clauses in collective agreements. This variety in practice is complemented by the wide range of entities that may both seek consultation and be sought for it. These entities range from more obvious examples, such as ministries, equality bodies, trade unions, and NGOs, to representatives drawn from rural areas and academic scholars. The selection of consultees is generally characterised by a laissez-faire approach; consulters are typically free to choose whom they wish to consult with. However, there are some notable exceptions to this, in which consultees are mandated by law.130 As has been shown, the question of obligation is one that can apply to the consultation process at any number of junctures. It is rare that the obligation is such that the opinions provided in the context of consultations are binding, resulting in de facto joint decision making. Rather, it is more common that the taking of notice is entirely discretionary, attended sometimes by a mere obligation to engage in consultation. In some of the consultative treatments dealing with the provision of comments and

126 Any conclusion reached or recommendation made by the Office of Equal Opportunities

Ombudsperson is not binding. The Supreme Administrative Court of Lithuania has even rejected the claim of the Office to shift the burden of proof to the defendant in administrative proceedings concerning sexual harassment. For the judgement, see Ministry of Social Security and Labour http://www.socmin.lt/index.php?294822287, accessed on 11 November 2009.

127 With respect to the operative aspects of Latvian equality policy, the role of consultation is limited to the mutual exchange of information and ideas. The Ministry of Welfare provides information on available resources and intended actions, while consultees attempt to draw attention to pressing needs. Although debates and ideas should in theory allow for the Ministry to make good decisions, they are disregarded in practice. Any consensus reached during course of consultation has no binding effect on the Ministry or any other actor.

128 This refers to Norway’s Working Environment Committees and employer consultation. Such consultations have no binding effect.

129 The social consultations mentioned above have no binding effect on the drafting of amendments to legislation. There is no legal instrument that regulates the methods of reaching social consensus in controversial issues.

130 See, e.g. Finland and Italy.


opinions on draft bills there is a requirement to explain a failure to take a consulted view into account.131 Whilst helpful to the decision-making process, consultation can impose significant resource demands on already stretched NGOs or interest groups. The effect of this has been referred to as ‘consultation exhaustion’. It is rare, however, for Member States to provide support for consultees. In Estonia, while some funds are designated under the state budget to support the activities of non-governmental organisations, financing is predominantly project-based. In Finland, some assistance is provided by the Equality Ombudsman. In Ireland, both the Inter-Departmental Committee and the Co-ordinating Committee are chaired and serviced by the Department of Justice. Germany also provides support personnel to the relevant Public Body’s Ombudsperson should the number of employees for that Public Body (under Federal law) exceed 1 000. 2.3 Monitoring Monitoring is essential to test the efficacy of proactive measures. Unlike an individual complaints model, which is concerned with a self-contained incident, proactive measures are programmatic and on-going. A process of monitoring and review is therefore necessary to assess whether a proactive measure is effective, to review its progress, and to readjust it if necessary. Monitoring of one sort or another is relatively widespread, with 18 Member States reporting that monitoring took place (Austria, Belgium, Finland, France, Germany, Iceland, Ireland, Liechtenstein, Luxembourg, Malta,132 the Netherlands, Norway, Poland, Romania, Slovakia, Slovenia, Sweden, and the United Kingdom). Successful monitoring generally requires the collection of gender statistics. There was a marked dichotomy between Member States who collected no statistics at all, and those which did. Thus, there is no systematic requirement to collect statistics in Bulgaria, Cyprus133, Denmark, Germany, Greece, Hungary, Italy, Lithuania, or Portugal. In Estonia, according to the Gender Equality Act, employers have the duty to collect statistical data concerning employment differentiated on the basis of gender and which allow, if necessary, the relevant institutions to monitor and assess whether the principle of equal treatment is complied with in employment relations. However, the Government has not yet adopted the pertinent regulation, and this requirement has thus not been implemented in practice. In other Member States, there are now quite comprehensive measures to collect gender-disaggregated statistics. In many countries, this takes place at a national level, in co-operation between the equality body and the general statistics-gathering process. For example, in Romania, the 2002 Act on Equal Opportunities stipulates that the National Institute for Statistics should support the activities of and cooperate with the National Agency for Employment in order to develop gender statistics and to implement the gender indices promoted by the European Commission. Sex-segregated data must be incorporated by the National Institute for Statistics into all documents, analyses, and assessments in relation to economic, social, and political activities. In

131 See, e.g. Estonia. 132 Malta’s equality body has monitoring powers, but note that there are no real proactive measures. 133 Although there is no statutory obligation to collect statistics, the government statistic department

collects various information mostly regarding the pay gap and the participation of women in the labour market. The results show that the pay gap between men and women has decreased significantly since 1995 (29 %) and keeps following a downward trend from 25 % in 2005 to 24 % and 22.8 % in 2006 and 2007 respectively.


Norway, thorough statistics are provided by Statistics Norway. Additionally, annual reports are generated by the Equality and Anti-Discrimination Ombud. In Poland, public statistics provide sufficient data to determine the areas of gender discrimination. In the field of employment, where special studies are conducted on a regular basis,134 those data are particularly detailed. In Finland, statistics on gender equality, including equality barometers, are routinely produced by Statistics Finland. In Ireland, as part of the objective of ensuring the implementation of the National Women’s Strategy, the Central Statistics Office along with all Government Departments and agencies has a general responsibility for ensuring that good-quality gender-disaggregated data is made available with regard to gender equality. Luxembourg has improved its collection of statistics during the last few years. This is mainly due to the action of the Ministry for Equal Opportunities which initiates projects in cooperation with the ‘Service central de la statistique et des études économiques’ (Central service on statistics and economic studies). However, it is not sufficient simply to collect data. It is also important to make use of statistics to assess and review progress of proactive measures. In some Member States, this link is not made. Thus, in Latvia, the implementation and results achieved under action programmes must be assessed each year and information must be delivered for approval to Cabinet of Ministers. However, although gender statistics are collected by the Central Statistical Bureau of Latvia, there is no statutory monitoring obligation. According to the expert, this is to be considered the most serious problem in Latvia – both policy and legislative measures are adopted in almost all cases without proper qualitative and/or quantitative research on the factual situation. Several Member States do, however, make an express link. In some contexts, the aim is to provide an initial diagnosis of the causes of gender discrimination. In Poland, the Department of Women, Family and Counteracting Discrimination conducted a project entitled ‘Monitoring of equal status of women and men’ between April 2006 and June 2007. The system was based on the analyses of official statistical data and results of reliable scientific research. It did not propose any solutions, but merely defined the problems. In other contexts, statistics are in practice used to assess the progress of proactive measures or the impact of future policies. Thus, under the Belgian Gender Mainstreaming Act of 12 January 2007, every member of the Federal Government is required to make sure that any statistics collated by the public bodies which s/he supervises135 are sex-segregated in order to facilitate gender assessment. Although there is no express requirement that these statistics be used to monitor gender-mainstreaming measures, statistics are in practice necessary. This is because, firstly, a gender-assessment note must be attached to any draft legislation or regulation, and it would not be meaningful without relevant statistics. Secondly, the Gender Mainstreaming Act provides that the Federal Government must submit an interim report (at mid-term) and a final report (at the end of each four-year term of office) to both Houses of Parliament. Statistics are in practice necessary for these reports. Similarly, in Finland, statistics and special studies are being undertaken in order to evaluate the gender equality policies of the last 10 years for the purpose of preparing the Government Report to be presented to Parliament in 2010. Parliament is expected to react to the information gathered for this Report. Indeed, the provisions on equality planning are based on the idea that statistics should reveal the need to review 134 For example, BAEL research (Research on economic activity of population) conducted by Main

Statistical Office since 1992 with the use of ILO methodology. 135 For example, the number of unemployed persons.


policies. In Luxembourg, the recent increase of assessment studies should facilitate review of existing provisions. As this is a new approach, no concrete results have yet been identified. Particularly important is the need to review proactive measures in the light of the assessment of existing measures. To do so, it is necessary not just to produce statistics, but also to create indicators and benchmarks, which set targets and enable progress to be assessed. In Austria, for example, an inter-ministerial working group is required to develop indicators in order to assess the effectiveness of the five-year National Action Plan on De-facto Equality of Women and Men, in particular on the development of full-time work and labour participation of women, the improvement in women’s incomes and the representation of women in leading positions. In Spain, the Government is required to evaluate regularly the effectiveness of the principle of equality in its policy areas, including the efficacy of empowerment measures.136 The supervision and monitoring of promotional measures is conducted by Equality Units belonging to each of the ministries, who gather the statistical information compiled by each Ministry. The process of review is well developed in Luxembourg, where an assessment has now been conducted of the latest ‘Plan d’Action National d’égalité des femmes et des hommes’ (National action plan on equality for women and men), covering the period 2006–8. The next plan will take the results into account in order to improve the impact of the national plan. Similarly, in the Netherlands, the equality policy, including proactive measures, is monitored biannually by the Social Cultural Planning Office, and the Central Office of Statistics. Statistics are collected and assessed by these bodies. Together, they publish the biannual Emancipatiemonitor.137 The Government itself also produces an annual ‘progress report’ with respect to measures mentioned in their main policy documents in the field of gender quality. In addition, once every four years the Government produces CEDAW reports about fulfilment of the Treaty. These reports are followed up very thoroughly by shadow reports of women’s rights organisations. These shadow reports have proven to be a valuable means to put issues on the political agenda. In some countries, there are monitoring obligations but they do not take the form of collection of statistics. There is a range of different methods of monitoring, some of them quite vague and general. In Finland, for example, the Equality Ombud monitors the Act on Equality in its entirety, but there is no obligation to report to the Ombud on measures taken by relevant bodies (employers, public bodies, educational institutions) and the Ombud has no special resources for monitoring equality planning. The result is that there are attempts to monitor some sectors and employers, but no systematic monitoring. In Germany, the Federal Ministry for Family, Senior Citizens, Women, and Youth (Bundesministerium für Familie, Senioren, Frauen und Jugend) monitors ‘voluntary commitments’ (Freiwillige Selbstverpflichtungen) entered into by various employers’ organisations. The Ministry publishes regular assessment reports, which have been criticised by women’s organisations for reflecting an overly sanguine view of the progress made with respect to proactive measures. In the Czech Republic, the national action plan for promoting equal opportunities is subject to a yearly assessment of action taken, which takes into account not only action taken at the governmental level, but also at the level of public authorities, trade unions, private bodies, employers, and others. One of the aims of the national action plan is to 136 Article 51(4) LOI. 137 Emancipatiemonitor http://www.scp.nl/content.jsp?objectid=default:20084, accessed 30 November

2009.


monitor and assess the effectiveness of applying the principle of the equal position of women and men. However, there is no obligation on any body to collect statistical data, nor even to review proper policy in light of the statistics. Similarly, in Liechtenstein, Article 18(2)(e) GLG enumerates as obligations for the Gender Equality Commission: the observance of the development concerning gender equality, the tracing of the adequate measures, and the periodic reporting about such practices to the Government. This is not, however, linked to systematically drawn statistics. In Slovenia, the responsible bodies are required to monitor the National Programme and periodical plans adopted for 2006–7 and 2008–9. However, there is no obligation to collect statistics. In the United Kingdom, monitoring is part of the positive duty of public bodies to pay due regard to eliminating unlawful gender discrimination and to promoting equality of opportunity. The same might be said of Germany, in which Public Bodies are required to collect information in order to compare the situation of male and female employees and to evaluate the measures taken so far with respect to the advancement of women in relation to the applicable Women’s Advancement Plan and finally, to remit this information to the relevant federal authority (usually the Ministry to which the Public Body belongs). There is no express obligation to collect statistics, but it seems unlikely that the objectives mentioned could be achieved without doing so. A different approach is to incorporate social partners at enterprise level into the monitoring process. Thus, in France, employers with at least 50 employees must present to the Works Councils, each year, a written report on the comparative situation of men and women in the enterprise. The structure of this report has been modified several times to improve its content. The employer must record the measures taken in the enterprise in the previous year towards attaining employment equality and provide an outline of the objectives for the year ahead. The report must compare men and women in terms of recruitment, training, qualification, pay, working conditions, and the balance between professional and private life, using relevant statistically-based indicators. The Government has also published models of reports to be used by the enterprises. As the French expert points out, this report is a fundamental tool to tackle the pay gap and discrimination. Without a shared and complete diagnosis of the equality situation, collective negotiations do not have any chance of succeeding and the content of the resulting collective agreements will be very weak. In other Member States, responsibility for monitoring lies with the equality body. In Iceland, enterprises and institutions with more than 25 employees must provide the Centre for Gender Equality with a copy of their Gender Equality Plan. The Centre for Gender Equality is responsible for monitoring gender equality developments in society (i.e. gathering information and initiating research). Gender-disaggregated statistics and information must be produced in the compilation of official economic reports on individuals and in canvassing interviews and opinion surveys. In addition, a distinction must be drawn between the sexes in the collection of data, data processing and the publication of information unless special circumstances (e.g. the protection of personal privacy) militate against such a course. In Ireland, the Equality Authority has the power to review the operation of the Employment Equality Acts 1998–2008 and may provide a report or make recommendations regarding any amendments to the Minister for Justice, Equality and Law Reform.138 The Authority is also obliged to carry out such a review if required by the Minister for Justice. In Norway, the Ombud monitors the duty to report found in Section 1a of the Gender Equality Act. In

138 Employment Equality Acts 1998–2004, Section 73.


Sweden, the equality body has a general duty to monitor compliance of the Act. To this end, there is a duty on any natural or legal person concerned to provide the information required by the EO. Should such a person not fulfil his or her obligations concerning active measures under Chapter 3, s/he may be ordered to fulfil them subject to a financial penalty. In practice, however, it is not clear that monitoring obligations are carried out as well as they should be. In France, for example, it seems that many enterprises do not compile the obligatory report on the comparative situation of men and women. An analysis of collective agreements concluded on equality also shows that most do not mention or use the comparative report. A similar pattern is evident in Slovakia. Here, state administrative bodies, which are authorized to adopt temporary compensatory measures, are obliged to monitor, evaluate and publish the adopted temporary compensatory measures in order to reconsider whether they should be further implemented. They are not required to collect statistics. Reports must be submitted on these issues to the National Centre for Human Rights – the national equality body. However, no reports were registered during the year 2008. Some information about the implementation of temporary balancing measures has been collected using questionnaires circulated by the Centre to the relevant state authorities. From answers in the questionnaires it may be concluded that the state administrative bodies themselves do not have a clear and unambiguous view of the legal institute of temporary compensatory measures or their application. This difficulty is exacerbated by the rather vaguely formulated obligation laid down by the Act on submission of reports to the Centre, which gives no exact definition of the parameters of such reports. The result is that state administrative bodies cannot not reasonably be expected to fulfil this legal obligation satisfactorily. 2.3.1 Confidentiality and data protection In general, there are no specific obstacles to the collection of gender statistics based on privacy or other legislation, as long as they are anonymous. For example in Belgium, legislation concerning the protection of privacy (mainly the Act of 8 December 1992) is no obstacle to the collation and publication of statistics by a public authority, provided that they do not lead to the identification of persons. The same can be said of Germany. Similarly, in Finland, information in the equality plans should not be of a nature to allow identification of single individuals. In Iceland, gender-disaggregated data must be collected unless special circumstances, such as the protection of personal privacy, argue against doing so. However, a key source of difficulty arises in respect of equal pay. Thus, in the Czech Republic, it is noted that the difficulty in identifying the gender pay gap arises in part from the fact that levels of wages are confidential and employers do not like to disclose what wages they pay to men and to women. Even employees are precluded from informing each other of the amount of their wages. This makes it very difficult to identify a gender pay gap at the enterprise or employment level. In Estonia, data protection issues have also been raised with regard to collecting data about wages. In Finland, confidentiality of pay information in the private sector is highly protected. Information on pay discrimination is also difficult to get, unless the person whose pay is to be compared with the plaintiff's is willing to provide it. An employee cannot receive such information from the employer. A representative of the employees may ask for it, but information on an individual employee may not be given against his or her will. In such cases, a representative of the employees may ask the Equality Ombud to ask for the information – a very cumbersome procedure.


Some Member States have dealt with this difficulty. One such example is Iceland. Workers may now, upon their choice, be permitted to disclose their wage terms (Article 19). However, in the opinion of the expert, this is not likely to be effective as it is left to the discretion of the employee to reveal such information. Someone who obviously enjoys better terms is not likely to disclose such information to someone who fares worse in a competitive environment. In Luxembourg, data may be processed if it is necessary in order to comply with legal obligations to which the controller is subject. This allows employers to collect sex-segregated statistics to comply with their obligations regarding pay equality. In Norway, an employer is generally not under an obligation to provide detailed information regarding pay to individual employees, but the enterprise must provide information about the pay difference in percentage between men and women. In Poland, there is a prohibition on mandatory collection of data on race, belief or philosophical or political opinions, as well as data related to ethnic origin, party or trade union membership, health, genetic code, addictions or sexual life, convictions and other decisions issued in court and administrative proceedings. There are, however, several exceptions from this prohibition of processing sensitive data.139 Since, for example, information about sex or age is not considered to be sensitive data, there is no obstacle to collecting relevant data in public statistics and to processing them in order to justify a course of action (e.g. undertaking of the positive action).140 In the National Programme of Activities on Women, statistical arguments have been frequently used, in order to prove women’s discrimination and the need to apply proactive measure to eliminate it. In the United Kingdom, a new provision in the Equality Bill currently before Parliament will, if it becomes law, invalidate clauses in contracts of employment which prohibit employees from disclosing their pay to colleagues. Employees who do make such disclosures are protected from victimisation. In addition, the relevant Minister will have the power to make regulations requiring employers to publish information relating to the pay of employees for the purpose of showing whether there are differences in the pay of male and female employees. Even if the Bill becomes law, it is not clear whether and in what form such regulations will be promulgated. 2.4 Enforcement Without the ultimate sanction of judicial procedures, there is a risk that proactive measures might become mere rhetorical gestures. A key challenge for proactive measures, therefore, is to devise appropriate means of enforcement. This has proved to be the most problematic aspect of proactive measures. Much depends on political goodwill and a sense of responsibility on the part of duty bearers, and when these are lacking, there is no easy solution. Several possible enforcement mechanisms were canvassed. One is to require regular reporting to Parliament, thereby relying on the political process to ensure that proactive measures are carried out. A second possibility is to give enforcement powers to relevant equality bodies. Such powers can include punitive sanctions, but more often involve engagement and oversight. A third possibility is to give the individual a right to complain, not about infringement of her own right, but about failure to take appropriate steps to fulfil a proactive measure. Procurement or contract

139 Among others, when the subject has given his/her written consent, the processing relates to data

necessary to pursue a legal claim. Collection is necessary to allow the data administrator to perform its duties related to employment of persons (within the limits defined by a legal act).

140 Or to submit statistical data to the court as evidence in a litigation case.


compliance, which may well be an effective means of enforcement, was not within the scope of this study.141 A final possibility is to offer financial incentives. For example, in France to promote equality, an ‘Equality Label’ was created in 2005. This Label is awarded to businesses with the best gender equality at work practices, and by the end of June 2009, 46 Labels had been awarded. 2.4.1 Reporting to Parliament The practice of reporting to Parliament is widespread among Member States. Seventeen Member States included such reporting mechanisms (Austria, Belgium, Estonia, France, Germany, Greece, Hungary, Iceland, Italy, Latvia, Liechtenstein, Luxembourg, the Netherlands, Romania, Slovenia, Spain and Sweden). The responsibility for providing the report can rest with the Government as a whole, with specific Ministries, or with a Parliamentary committee. Thus, in Belgium, the Gender Mainstreaming Act provides that the Federal Government must submit an interim report (at mid-term) and a final report (at the end of each four-year term of office) to both Houses of Parliament. Similarly, in Spain, the Government is required to issue regular reports on the effectiveness of the equality measures that must be submitted to Parliament. Germany also exhibits a similar trend, in which the Government is required to report to Parliament on the situation of women and men in Public Bodies and provide a description of the measures taken to improve the situation. In Iceland, this task lies with the Minister of Social Affairs and Social Security, who is required to present a motion for a parliamentary resolution to Parliament on a four-year gender equality action programme after having received proposals made by the ministries, the Centre for Gender Equality and the Gender Equality Council. In Italy, the Minister of Labour, together with the Minister for Equal Opportunities, on the basis of the annual reports of the Equality Advisers Net and of the EONC, must report to Parliament every two years on the state of enforcement of gender equality law. A similar report is also annually presented to Parliament by the Minister of Industry and Trade as regards the enforcement of equal opportunities principles in the sector of entrepreneurship. In Poland, the enforcement of the projects, aimed at promoting gender equality, has been conferred on the Department for Women, Family and Counteracting Discrimination. In Greece, the Ombudsman’s annual report is submitted to Parliament; while the National Human Rights Commission’s annual report is submitted to Parliament, the government and other competent authorities. In Cyprus, by contrast, responsibility lies with the Parliamentary Committee for Equal Opportunities, established in 2006. Its aim is to monitor government policies and actions on the issue of equal opportunities between men and women and support the implementation of the National Action Plan for Gender Equality, the NMWR and the efforts of governmental and NGOs which promote gender equality. There is no obligation to submit a report to Parliament as a whole. A similar approach is adopted in Hungary, with the parliamentary subcommittee on ‘Social Equality of Women and Men’, belonging to the permanent committee on ‘Human Rights, Minority, and Civil Affairs’, monitoring the implementation and impact of relevant laws. In Ireland, there is no duty to report to Parliament but there is a duty on the Equality Authority to produce Annual Reports to the Minister for Justice, Equality and Law Reform detailing the performance of its measures including any equality reviews it has undertaken.

141 See C. McCrudden ‘Buying Equality’, European Anti-Discrimination Law Review No. 8 2009,

p. 17.


The effectiveness of the reporting mechanism is dependent on the seriousness with which it is regarded by Parliament or the relevant Ministry and whether further action is taken. The risk is, as identified by the Latvian expert, that the reporting procedure might become no more than a formality. Thus, although the responsible Ministry should report yearly to the Cabinet of Ministers on the state of implementation of envisaged measures, in practice, the Cabinet of Ministers accepts facts on the state of implementation rather than analyses whether implementation of measures envisaged were effective and the desired result has been achieved. Indeed, proactive measures taken by Latvia are, in the view of the expert, insufficient and ineffective, primarily due to the lack of political will, which in turn leads to an absence of proper legal regulation and of financial resources. There is little information on how this process works in other jurisdictions: for example, the Polish report points out that no information is available on how the Department for Women, Family and Counteracting Discrimination carries out its enforcement function. This is not to say, however, that the process itself is not useful. As the Liechtenstein expert comments, there is no single magic formula; instead, every step counts. 2.4.2 Equality bodies A second possibility is to give enforcement powers to equality bodies or other similar institutions. Such powers are given to the equality body in eleven Member States (Bulgaria, Estonia, France, Ireland, Hungary,142 Lithuania, Norway, Romania, Sweden, Slovenia and the UK). The ideal model in this context would be a pyramid of enforcement, whereby the first response to non-compliance would be for the equality body to initiate a process of discussion and negotiation. If this is not successful, the recalcitrant respondent could be subjected to an order to comply issued by the equality body. Only if this further step fails do fines or other judicially enforced sanctions come into play. The Swedish Discrimination Act 2008 follows a pattern of this sort. The Equality Ombudsman, which is responsible for supervising compliance with active measures under the Act, is required to try in the first instance to induce those to whom the Act applies to comply with it voluntarily. To this end, it has powers to require anyone who is subject to take active measures, to provide information about the circumstances of their relevant activities, to give the Ombudsman access to workplaces and other premises where the activities are conducted for the purpose of investigations, and to attend discussions with the Ombudsman.143 Further powers are available if voluntary compliance is not forthcoming. In such circumstances, the Ombudsman may order a person who does not fulfil her or his obligations concerning active measures to fulfil them subject to a financial penalty. Such orders are issued by a special Board against Discrimination on application from the Equality Ombudsman. They can also be directed towards the State as an employer or as the entity responsible for educational activities. If the Ombudsman has declared that she or he does not want to apply to the Board for a financial penalty to be ordered, a central employees’ organisation with respect to which the employer is bound by a collective agreement may make an application concerning active measures in working life under Chapter 3, Sections 4-13. A slightly different permutation is found in the UK, although the first, conciliatory step is less explicit. Here the aim is to induce initial compliance through 142 Although the degree of this authority is limited, in that the ETA is only competent to review the

compliance of public (or semi-public) employers with the relevant Equal Opportunity Plan. 143 Failure to comply with such orders are themselves subject to financial penalties.


the more general processing of auditing public bodies. Inspectorates with responsibility for checking that public bodies generally comply with their responsibilities are also charged with ensuring that equality duties are observed. Thus the Code of Practice points out (Paragraph 4.5) that inspectorates ‘with a broad role will need to ensure that the [equality] duty becomes an integral part of the inspection/audit process, built into their inspection regimes and informing their judgements on what constitutes good performance’. In particular, where appropriate, they will need to review inspection and auditing methods and performance indicators, to ensure that they meet the duty and enable judgements to be made as to whether public authorities are complying with the duty; advise public authorities on developing effective gender equality schemes, action plans and monitoring arrangements; identify and disseminate best practice in respect of the gender equality duty; improve research surveys and data collection in order to provide useful data for public authorities to consider when analysing their performance of the duty, and to improve accountability to the public.’ As a second stage, the Equality and Human Rights Commission can issue an enforcement notice in respect of failures by public bodies to comply with the specific duties set out in regulations. These can ultimately be backed up by judicial proceedings. A similar practice is seen in Hungary, in which the equality body – the ETA – exists as a public administrative organ charged with the task of investigating public (or semi-public) employers. The ETA can investigate compliance with an Equal Opportunity Plan, and, if it concludes that a public employer has failed, order compliance. It is also empowered to apply sanctions in circumstances where it is of the view that the Equality Opportunity Plan has not been complied with. However, any power of enforcement held by the ETA is limited to situations in which there is an existing Equality Opportunity Plan that has been developed and applies to the public employer being considered. In Italy, by contrast, the focus is on the first stage, namely inducing voluntary compliance. Thus, the Equal Opportunities National Committee, while having no powers of enforcement, can propose solutions to collective disputes, by helping the parties to carry out positive action plans targeted both at the removing discrimination in case of discriminatory practices and of mere imbalance arising from social attitudes, behaviour and structures and at achieving equal opportunities for female employees through the co-financing of positive action plans. This function, however, has been seldom realized (on the basis of available data). National and Regional Equality Advisers can also propose a conciliation agreement as an incentive to employers to avoid administrative sanctions. This would require the person responsible for the discrimination to set a plan to remove it within 120 days. If the plan is considered fit to remove discrimination, on the Equality Adviser’s demand, the parties sign an agreement which becomes a writ of execution through a decree of the judge. In other cases, enforcement powers of an equality body may in themselves consist only of a duty to report or receive reports, thus overlapping with their monitoring responsibilities. For example, in Slovenia, pursuant to the Act on Equal Opportunities for Women and Men, Ministries and government offices must report on the implementation of the activities under their responsibility to the Office for Equal Opportunities two months before the expiration of a two-year periodic plan. On this basis, the Office draws up a report on the implementation of the National Programme in which it states the measures (including proactive measures) and activities that have been carried out. Every two years it reports to the National Assembly.


A more targeted, and possibly more effective approach, is through a power to review employers’ obligation to produce annual equality plans. A good example is found in Norway, where, as was noted above, the Gender Equality Act imposes an obligation on private and public companies as well as Ministries to provide annual reports/budgets, providing an account for the actual state of affairs as regards gender equality in the enterprise. An account must also be given of measures that have been implemented and measures that are planned to be implemented in order to promote gender equality and to prevent differential treatment in contravention of the Act. Enforcement of these duties lies with the Equality and Anti-Discrimination Ombudsperson which reviews and evaluates the reports. A different approach is to provide for a network of institutions and responsible persons for the enforcement of equal treatment legislation. Thus in Austria, the Equal Treatment Act provides for ombudspersons (up to seven in every Ministry), working groups on equal treatment questions within the ministries and universities, as well as an inter-ministerial working group chaired by the Minister for Women’s Affairs. 2.4.3 Individual complaints Several reports point to the need to retain some initiative for individuals within a proactive model. As the report for Iceland pointed out, proactive measures might better serve the objective of achieving real and substantive equality if tools were granted to the victims themselves. However, most Member States have no such possibility. The ground most frequently given is that proactive measures by their very nature do not generate individual rights. For example in Italy, several judgments have tackled the problem of a public competition in which the rule that one third of members of the commission must be female was not respected. Although judges underlined the importance of this provision, which can be considered a proactive measure, they did not recognize that the workers had an actual and autonomous interest on the basis of which they could complain in the court in respect of its violation.144 A handful of Member States do allow individuals to bring complaints where proactive measures have not been fulfilled: only Belgium, Bulgaria, Portugal and Slovenia identified such measures. A rather limited and undeveloped right might also be said to persist in Hungary, where individual complainants can initiate an investigation with respect to a public (or semi-public) employer’s failure to adopt an Equal Opportunities Plan, but not for the lack of implementation of an adopted plan. In other words, employers might be made responsible for not adopting an Equal Opportunity Plan at all. However, if one has been adopted, no one can bring them to any authority or court for not implementing the plan, since the plan does not establish individual rights. The limitations of such an individual complaint are patent. Judicial review is particularly rare, applying only in Italy, Portugal, Slovenia, Ireland, the UK, and, in a very limited way, Hungary.145 Bulgaria permits individual complaints to be used for the enforcement of proactive measures which are

144 Tribunal of Siena 16 June 2006, in Rivista Critica di Diritto del Lavoro 2007, p. 136 and Consiglio

di Stato 6 June 2002 n. 3184: http://www.giurdanella.it/pdf.php?id=6542, accessed on 23 July 2009. 145 In this respect, judicial review in Hungary is only available in relation to enforcing provisions of a

collective agreement. That said, it is extremely rare for collective agreements to include provisions relating to proactive measures.


formulated as obligatory, but no case law exists so far.146 In the UK, individuals cannot complain about failure to comply with the specific duties but can challenge the failure to comply with the general duty (which could be evidenced by failures in respect of the specific duties) by way of judicial review. A challenge by way of judicial review would consist in an assertion that a public authority acted unlawfully by failing to comply with its general duty. The complainant would be anyone with a sufficient interest (potentially anyone affected or potentially affected by the failure, as well as representative bodies including, but not limited to, the EHRC). The remedy is generally in the form of a declaration that the public authority has acted unlawfully, though quashing and mandatory orders are available. A form of judicial review is also available in Germany, but its application is extremely limited. As has already been mentioned, Public Bodies having more than 1 000 employees appoint an Equality Ombudsperson who monitors the implementation of the employer’s Women’s Advancement Plan. Although the implementation of this plan itself is not judicially reviewable, the failure to accord the Ombudsperson her legally entitled rights with respect to participation in the process of formation and amendment is. Where individual complaints are possible, they often occur indirectly. In Belgium, for example, the programmatic nature of the provisions of the Gender Mainstreaming Act rule out the possibility of any direct effect on which individual complaints could be grounded. In the worst case (i.e. should a new regulation or piece of legislation induce blatant gender discrimination that could have been avoided through a proper gender mainstreaming survey), failure to comply with the Act of 12 January 2007 might serve as an argument in an application for annulment by the Conseil d’Etat/Raad van State or the Constitutional Court, relying on the Gender Act or Constitutional provisions such as Article 10. In a more informal way, an interested individual or organisation might complain to the Institute for Equality of Women and Men or the Council about a measure which has been proposed by a Minister.147 This might lead the Institute or Council to intervene with the concerned authority, in reliance on the Gender Mainstreaming Act, in an attempt to prevent the measure being instituted. In Italy, the relationship between individual complaints and proactive measures operates in the reverse direction, that is, positive action can constitute a remedy for a successful individual complaint. Thus, positive action can be part of an order of the judge or of a conciliation agreement between the employer and an Equality Adviser, which becomes a writ of execution with a decree of the judge. More promising is the potential for trade unions to complain in these circumstances. As we have seen, in Sweden, if the Ombudsman has declared that she does not want to apply to the Board for a financial penalty in relation to a breach of the duty to take proactive measures, an application for enforcement may be made by an employees’ organisation which has a binding collective agreement with the employer in question.148 In France, the social partners have a duty to include measures promoting gender equality in their collective agreements. If proactive measures defined by a collective agreement are not respected by the employer, the employee and/or trade unions have a right to claim enforcement of these measures before a

146 The only area in which complaints have been used so far has been to contest admission quotas in

universities on the grounds that they do not comply with the requirement that they be justified and temporary in nature.

147 Given the propensity of most Ministers to herald their brainchildren in the media even before they are adopted.

148 Chapter 3, Sections 4–13.


court.149 If the tribunal finds that the measures are mandatory, according to the collective agreement, it will order the employer to apply them and can also order some remedies. 2.4.4 Effectiveness There is little experience with the workings of enforcement powers in the context of proactive measures. In Bulgaria, for example, the Commission has broad competences in the field, but there have been no cases of enforcement of positive measures for gender equality. More detailed evidence of the operation and effectiveness of compliance mechanisms is found in Norway, where the Ombud has used its powers to carry out several systematic reviews since the introduction of the report and activity duty in 2003. In 2004 and 2005 several private enterprises were investigated. This revealed that most companies had fulfilled their obligation to report, but that the reports were mainly brief and superficial. This was followed in 2007 by a review by the Ombud of fifty municipalities. This revealed that all 50 of the municipalities had handed in their reports to the Ombud and the majority had some sort of provision regarding both the report and activity part following the requirements of the Gender Equality Act. However, 7 of the municipalities had no information on gender equality issues at all. The Ombud approved 37 reports, but 13 municipalities did not get their reports approved and were informed that the Ombud would return and perform a review of the improved reports. Moreover, the Equal Pay Commission has stated that the quality of the equal pay figures from the municipalities was not sufficient in quality to allow the Commission to draw conclusions as to the factual pay difference between men and women. One of the particular difficulties is that proactive measures are themselves ill-defined and therefore it is not easy to determine whether they have been breached. This can undermine the most sophisticated of enforcement mechanisms. This issue was identified in several Member States. In Sweden, as we have seen, the system of enforcement is carefully constructed. However, the rules on active measures in the 2008 Discrimination Act, although by nature obligatory, are quite imprecise and, in the opinion of the expert, the sanctions provided cannot be said to be very efficient. Similarly, in Romania, responsibility for enforcement of the 2002 Act on Equal Opportunities lies with the Equality Agency. The main function of the Agency is to ensure active and visible gender mainstreaming in all national policies and programs. The 2002 Act stipulates that the Agency’s President must submit activity reports and analyses to the Parliamentary Commissions on Equal Opportunities on a half-yearly basis. Of particular importance is the fact that such analyses and reports must target the ways in which budgetary and extra-budgetary allocated financial resources are utilised for implementing equal opportunities legislation. Therefore, at least on paper, the declared nature of such proactive measures is very generous. However, as the expert notes, proactive measures designed to combat discrimination and achieve equality as provided for by the 2002 Act lack a clear strategic content organised alongside concrete measurable objectives. Thus, the sophisticated enforcement powers operate in the complete absence of defining any concrete measurable objectives designed to assess the extent to which proactive measures have been implemented. Consequently, both the proactive measures’ nature, and their successful implementation, are very difficult to assess.

149 An industrial tribunal if it is an individual claim, a civil tribunal if it is a trade union claim.


Whether or not success has been achieved can itself be controversial. In Germany, the most recent monitoring report showed an increase of women’s education, a rise of women in technical studies, in apprenticeships and among the self-employed. It also noted a sharp increase of company measures for reconciling work and family life. However, women's NGOs argue that too little progress has been made. As the expert points out, the employment rate of women with children under the age of five is only 44 %, thus putting Germany in the lowest range in a Europe-wide comparison. Women are still rare in decision-making positions, and the gender pay gap persists. Moreover, the report did not contain a systematic analysis of measures taken within companies to promote women (Frauenförderpläne), which constitutes a main pillar of all (private) proposals for a law on gender equality in the private sector. According to a recent study by the Research Institute of the Federal Agency for Employment, 90 % of German enterprises have not concluded an internal agreement (with the Workers’ Council) on promoting women. The 10 % of enterprises which have such programmes are big companies; they employ about 20 % of the female workforce. Women NGOs also contest whether proactive measures are indeed promoting equality. They argue that measures for reconciling work and private life all too often focus on increasing women’s flexibility instead of on gender equality. In their view, such measures are responsible for women’s interrupted work biographies and for leading them into the ‘trap’ of part-time work, which creates serious obstacles for their careers and future pensions. The absence of sufficiently well-developed enforcement mechanisms in most Member States undermines the potential of proactive measures to achieve their aims. In Estonia, for example, the report notes that while there is a relatively comprehensive legislative framework to facilitate the promotion of gender equality by proactive measures, a number of legislative requirements have remained declaratory in nature or have not yet been fully implemented in practice. Four specific areas can be identified in which little or no action has been taken. The first area concerns the duties of the Minister. There is now a legislative power to enact specific measures to compensate for disadvantages experienced by persons belonging to one sex and to promote gender equality. The law also establishes duties on different types of actors to promote gender equality as well as requiring impact assessment upon planning, implementation and assessment of national, regional and institutional strategies, policies and action plans. Moreover, the Minister of Social Affairs is to make further recommendations concerning performance of these obligations. However, these duties have not been enforced and there is no information suggesting that they have been carried out in practice. Secondly, the Gender Equality Council, whose task is to approve the general objectives of gender equality policy, is yet to be established. Thirdly, the Gender Equality and Equal Treatment Commissioner has the duty to take measures to promote equal treatment and gender equality, analyse the impact of legislative acts on the position of men and women in the society, and accordingly make proposals to the Government, government agencies, local governments and their agencies for amendments to legislation. These powers have not been implemented in practice. It is hardly helpful that while the competences of the Commissioner have been expanded, the Commissioner has not received any additional funds to effectively carry out her tasks. Fourthly, the GEA envisages that employers will have duties to promote gender equality. However, thus far state activity in this respect has been limited to raising awareness of employers and employees on the basic concepts of gender equality. Furthermore, the State is yet to adopt implementing acts to collect


gender-specific statistics which would help to monitor the gender equality situation in employment. Similarly, the Finnish expert notes that the downside of the Finnish tradition of proactive-oriented gender equality policies is that the resources allocated to such policies are scarce. This is especially true of gender mainstreaming. Moreover, sanctions to enforce the various proactive measures that many actors have under the Act on Equality are toothless or non-existent, and the resources for monitoring equality plans are insufficient. The result is that the assumption that gender equality policies have evolved from legal formal equality to positive measures and further to gender mainstreaming, and from less to more effective measures, does not accurately capture the developments in Finland. A weak emphasis on anti-discrimination may also lead to a decline in the awareness of the need for effective positive measures. The Irish report was even more pessimistic. Given the economic difficulties over the next few years, it considered that there will be reduced state resources allocated to the promotion of gender equality. Indeed the Equality Authority has already experienced a significant decrease in its funding. Because of this, the Irish expert is of the view that the safest protector for the individual for the foreseeable future is the complaints-led model. Greater success is achieved when legal obligations are enforced through collective bargaining structures. In France, the traditional model of enforcement is completed by a model based on collective actors and collective bargaining. Since 2001, social partners have had to negotiate to define measures to promote gender equality. These obligations have been intensified by the 2006 law with the aim of reducing the gender pay gap by the end of 2010. As we have seen, if the proactive measures defined by a collective agreement are not respected by the employer, the employee and/or trade unions have a right to enforcement by the courts. The last annual report on collective bargaining confirms a slow increase in the number of collective agreements dedicated or referring to equality, but the subject of equality is still relatively marginal compared with the traditional topics of collective bargaining. 150 At branch level, 19 specific agreements on equality were concluded in 2008 (against 9 in 2007 and 1 in 2006) and 34 agreements refer to equality (out of a total of around 1 117 agreements concluded in 2008). At enterprise level, 1 235 agreements referring to equality were concluded, against 1 076 agreements in 2007, on a total of 27 100 agreements. Concerning the content of the agreements, the report distinguishes between 3 categories of agreements. The first category of agreements (representing 1/3 of the specific agreements, and 2/3 of the general agreements) are merely formal: they simply recall the principle of non-discrimination and declare their willingness to respect the law, without specifying any concrete measures. The second category of agreements (half of the specific agreements and 1/3 of the general agreements) recall the principles of non-discrimination and propose at least one concrete measure which is, in most cases, the neutralisation of the effect of maternity leave on wages (which is in fact a legal obligation). Some agreements specify some objectives in terms of career development and recruitment. Only the third category (23 specific agreements and no general agreement) tries to take into account the structural causes of gender discrimination and adopt various measures on recruitment, promotion, access to training, access to part-time work, parental measures, etc.

150 M. Rabier, ‘Analyse du accords d’entreprise portant sur l’égalité professionnelle entre les femmes

et les hommes signés depuis la loi du 23 mars 2006’, in Ministère du travail, La négociation collective en 2008, Bilans et rapport, June 2009, pp. 423–59.


In countries without a well-developed collective bargaining structure, this means of enforcing proactive measures is conspicuously absent. Thus, as the Romanian expert pointed out, Romania’s ability to enforce proactive measures is limited by the fact that industrial relations in Romania, as a new Member State, seem to be less developed than in the old Member States. Institutionalized mechanisms of workplace industrial relations in Romania are established only in a relatively small number of large enterprises and both the structures and the actors for formal bargaining and social dialogue are lacking. V CONCLUSION The introduction of proactive measures into the arena of gender discrimination holds much promise, as well as many new challenges. The architecture of proactive measures is increasingly visible within Member States’ provisions for gender equality, and there is a continuing and dynamic process of development and appraisal. However, structures on their own are not sufficient to ensure that proactive measures are implemented and have an impact. The greatest challenge remains that of creating appropriate incentives, sanctions and mechanisms for accountability to ensure that elaborate structures do not simply conceal apathy or proceduralism. The study suggests that political commitment and goodwill, together with the active involvement of stakeholders, particularly trade unions and other representatives of those affected, are essential for success. The danger remains that the location of proactive measures on the borderline between law and politics makes it appear that fulfilment of such measures is discretionary or optional. The ultimate challenge is therefore to ensure that proactive strategies are based on a recognition that equality is a fundamental right, not an optional policy.


Annex I

Questionnaire

European Network of Legal Experts in the field of Gender Equality Report: Making Equality Effective: The role of proactive measures

Background The aim of this report is to consider how to make the achievement of real and substantive equality between men and women more effective. There are at least two reasons why existing approaches may have limited effect. The first has to do with the definition of discrimination, which may be limited to a simple requirement that women be treated the same as similarly situated men (the formal equality approach1). The second has to with models of enforcement. Traditional models of enforcement, which depend on individual complaints, have generally had limited success. This has led to the introduction of new models, whose aim is to achieve equality for women through measures which go beyond individual complaints. Some of these widen the complaints-led model to permit collective complaints. Others go further and place the initiative on public bodies, employers, and others to tackle discrimination and promote equality even in the absence of a complaint. The EU itself has given particular priority to such measures, both in Article 2, which highlights equality between men and women as a common value of the EU, and Article 3 (2), which specifically states that the Union shall promote equality between women and men. A variety of such models has been introduced both in member states and in the EU itself. These measures are known by many different names, including mainstreaming, proactive measures, positive action and others. Because of the confusion in terminology, the term ‘proactive measures’ is used in this study to designate the alternatives to a complaints-led model. This is because, under whatever name, the essence of these models is that they are forward-looking, requiring bodies2 to take the initiative rather than merely responding to complaints. The idea of this study is to capture the many types of practices in member states which are proactive in that they aim to change existing practices, scrutinise future practices for their impact on women, or implement express policies to further equality for women. It aims to investigate the nature and extent of such proactive measures in member states, and, so far as the context allows, point to ‘best practice’ models for future development. The ‘complaints-led model’ The traditional model of enforcement of anti-discrimination law has generally relied on individual victims to bring complaints against alleged perpetrators. This ‘complaints-led’ model, while important, is limited in many respects. It leaves the

1 For example, EU equality law generally takes a symmetrical view of equality. Measures which

favour women in order to redress pre-existing disadvantage are permitted but only as an exception to the overriding notion of equality.. Please note that there will be a separate report on concepts of equality.

2 The term ‘bodies’ is used here to include a wide variety of actors, including public authorities, member states, civil servants and other government officials, equality bodies, unions, employers and others who are in a position to bring about change.


initiative to the individual victim, which may be burdensome and costly, with the result that there is no comprehensive enforcement and many breaches go unremedied. In addition, remedies are retrospective, and often individualised. Even when the complaint can be initiated by a union or a class, this model remains limited. There is no direct call for forward-looking, or proactive measures, which might bring about more far-reaching change. Moreover, the complaints-led model only addresses discrimination where there is an identifiable perpetrator (which might include the legislator) who can be proved to have breached the law. In fact, the inequality between men and women has causes which go well beyond individual prejudice, and require structural and institutional change. ‘Proactive’ measures in the field of gender equality As a result of these limitations but also for other reasons, like the fact that there is little litigation at all in this field, attention has shifted to the possibility of alternative and complementary means of making equality more effective. These focus on measures to promote or achieve equality between men and women. Rather than being initiated by individual victims against individual perpetrators, such approaches place the responsibility on bodies, such as public authorities or employers, who are in a position to bring about change, whether or not they have actually caused the problem. Such measures are generally future oriented. For example, they could require a body – to take the initiative in eliminating existing unlawful discrimination; – to scrutinise and change existing practices which perpetuate or cause

discrimination; – to assess all new policies for their impact on gender equality; – to create specific policies to promote gender equality.

There is a great variety of measures of this sort in the member states. For example, – some are based in policy measures (soft law) and others in legislation (hard law); – some are enforceable through judicial mechanisms, and others are not; – some include detailed requirements to conduct impact assessments of future

policies, monitor change and produce reports, while others are left to the discretion of policy-makers;

– some involve stakeholders in different capacities and other do not. Generally speaking, these measures go beyond a formal equality model, recognising that change would not be achieved if women were simply treated in the same way as men. They aim instead to change practices and policies, and this might include asymmetric approaches, treating women differently to men in prescribed situations. An important dimension of proactive change is to address the way in which women’s continued responsibility for child-care affects their ability to participate fully in the public sphere, including paid work, decision-making positions and education. Proactive ‘family friendly’ measures need to be carefully assessed to ensure that they achieve substantive change and encourage shared parental responsibility, rather than reinforcing women’s role. We appreciate that some of this material might be difficult to access, but we hope experts will produce what information there is. We recognise that this is a relative uncharted area, and that much of the material relating to proactive measures may take the form of practice and policy which is not


documented or difficult to find. At the same time, this shows how important it is to bring together the differing practices happening in member states and to consider future directions. Please feel free to include in the final question any material which does not fit neatly into the questions. We very much appreciate your efforts. Part I: The complaints-led model (Please feel free to draw on material you have already submitted for the purposes of our report on Gender Equality Law in 30 European countries.) 1) Please give a brief summary of the procedure in your country for an individual to

bring a complaint of gender discrimination in breach of the EU gender equality law as implemented in your country.

2) If statistics are available, can you please give statistics for the numbers of gender

discrimination claims to the bodies specified, including the numbers of successful claims (number and percentage if possible) and the nature of the remedy (damages (how many cases or what proportion of cases), reinstatement (how many cases or what proportion of cases) etc) in the most recent year for which statistics are available. (If it is easy to produce trends over the last five years, please do so.)

3) Please point out any significant modifications to the complaints-led model: For

example i. Are class actions permitted in the gender discrimination field?

ii. Is action permitted even without identifying a concrete victim? iii. Does the equality body have standing to bring claims on behalf of individuals?

If so in what circumstances? iv. Do trade unions, NGOs or other similar bodies have standing to bring claims?

v. Does the equality body investigate or adjudicate the complaint itself? 4) What in your view are the strengths and weaknesses of the complaints model as

experienced in your country? Part II: Proactive measures: 5) Does your country have any proactive approaches to combating discrimination

and achieving equality? Please give a brief general description. 6) Who is responsible for taking proactive steps? i. Public bodies (please explain which bodies this might cover)

ii. Trade unions iii. Private employers (excluding measures which are instituted purely

voluntarily) iv. Private bodies with public functions (please explain what functions this might

cover) v. Private service providers

vi. Others (please specify function) 7) What is the nature of the proactive measures? Do these cover: i. Eliminating unlawful discrimination, even in the absence of an individual

complaint?


ii. Assessing all existing policies and all new policies to determine their impact on gender, and modifying them if necessary? In this area, is there a ‘screening’ mechanism, whereby only policies likely to affect gender are assessed in this way?

iii. Designing new measures to promote gender equality? iv. Assessing gender pay gaps and taking steps to redress any pay discrimination

revealed? Is there for example an obligation to screen job evaluation schemes? v. Other. Please give details in each case. vi. If it is possible to identify what in your view is a ‘best practice’ model, please

do so. 8) What is the source of any proactive measures? Are they found in i. Statute ii. Constitution iii. Policy documents iv. Other. Please give details in each case. v. In each case, please specify the extent to which these measures are obligatory

or discretionary. (For example, in the UK, a public body must ‘pay due regard to the need to promote equality of opportunity’.)

9) What are the aims of these policies? Would you regard them as going beyond

formal equality? Are the aims specified? 10) Are the responsible bodies required to monitor their policies? i. Are they required to collect statistics initially to determine the nature of the

problem? ii. Are they required to collect statistics to determine the effect of any measures

on addressing the problem? iii. Are they required to review the measures in the light of the statistics? iv. Are there any confidentiality/data protection issues which might make the

collection of statistics for monitoring purposes difficult?

11) Is the responsible body required to consult others in relation to its proactive measures? i. Who is it required to consult with?

ii. How are consultees chosen? iii. Is there any support given to those who are consulted? (e.g. capacity building) iv. What is the role of consultation?

1) To give information 2) To gain information 3) To gather ideas which the body will then consider, but is not bound to

follow 4) To have a binding effect on decision –making?

12) Content of the measures: Please give some examples of the type of measures

which have been used. For example: i. Is there a general policy commitment to ‘mainstreaming’?

ii. Are there quotas or set numbers of places which are set aside for women in 1) Employment 2) Education


3) Legislature 4) Other decision-making bodies iii. To what extent are ‘family friendly’ measures included in mainstreaming or

proactive measures? So far as you can, please specify the nature of such measures (e.g. child-care, flexible working, parental leave). Please also identify, where possible, the extent to which such measures are limited to mothers, or are (also) available to fathers. Where statistics are available, it would be helpful to know what the take-up of family friendly measures is among fathers.

13) Enforcement: How are these measures enforced? i. Is there a duty to report to Parliament, a committee in Parliament or

equivalent? ii. Does the equality body have enforcement powers? If so, what are they? iii. Do other audit bodies have enforcement policies? Please specify. iv. Is there a right of individual complaint in case that the measures are not

enforced? If so, is it to a court, tribunal, equality body, ombud or other? (Please specify)

v. Is there judicial review of such measures? If possible, please specify who the complainant is, what the standard of review is, and what remedies are available.

14) Other: Please include any further comments which you think may be relevant to

this study.


Annex II

Bibliography

F. BETTIO & A. VERASHCHAGINA Gender segregation in the labour market: Root causes, implications and policy responses in the EU European Commission’s Expert Group on Gender and Employment (EGGE) European Commission Directorate-General for Employment, Social Affairs and Equal Opportunities March 2009.

EUROPEAN COMMISSION Equality Between Women And Men — 2009 Report From The Commission To The European Parliament, The Council, The European Economic And Social Committee And The Committee Of The Regions (Com(2009) 77 Final).

EUROPEAN COMMISSION Manual for Gender Mainstreaming 2008, available at http://ec.europa.eu/social/main.jsp?catId=421&langId=en&furtherPubs=yes

EUROPEAN NETWORK OF LEGAL EXPERTS IN THE FIELD OF GENDER EQUALITY Legal Approaches to Aspects of the Reconciliation of Work, Family Life in Thirty European Countries European Commission, Directorate-General for Employment, Social Affairs and Equal Opportunities, Unit G/2 August 2008, available at:

http://ec.europa.eu/social/main.jsp?catId=641&langId=en&moreDocuments=yes S. FREDMAN Discrimination Law (Clarendon Law Series, Oxford University Press

2002. S. FREDMAN ‘Changing the Norm: Positive Duties in Equal Treatment Legislation’,

[2005] 12 Maastricht Journal of European and Comparative Law pp. 369-398. S. FREDMAN Human Rights Transformed: Positive Duties and Positive Rights Oxford

University Press 2008. S. FREDMAN, ‘Reforming equal pay laws’, [2008] 37 Industrial Law Journal

pp. 193-218. S. KOUKOULIS-SPILIOTOPOULOS ‘Gender equality in Greece and effective judicial

protection: issues of general relevance in employment relationships’, Neue Zeitschrift für Arbeitsrecht Beilage 2/2008 pp. 74-82.

S. LAULOM ‘Gender Pay Gap in France’ European Gender Equality Law Review 1/2009 pp. 5-15, available at: http://ec.europa.eu/social/main.jsp?catId=641&langId=en&moreDocuments=yes

S. MAZEY Gender Mainstreaming in the EU London European Research Centre 2001. C. MCCRUDDEN ‘Advice to a Legislator on Problems regarding the Enforcement of

Anti-Discrimination Law and Strategies to Overcome Them’ in: T. Loenen & P.R. Rodrigues Non-Discrimination Law: Comparative Prespectives pp. 295-312 Kluwer Law International 1999.

C. MCCRUDDEN Buying Social Justice: Equality, Government Procurement, and Legal Change Oxford University Press 2007.

C. MCCRUDDEN ‘Buying Equality’, European Anti-Discrimination Law Review No. 18 2009 p. 17.

M.A. POLLACK & E. HAFNER-BURTON ‘Mainstreaming Gender in the European Union’, Journal of European Public Policy 2000 7:3 Special Issue pp. 432-56.

T. REES Mainstreaming Equality in the European Union Routledge 1998.

EUROPEAN NETWORK OF LEGAL EXPERTS IN THE FIELD OF GENDER EQUALITY